This patch adds datapaths of interest support where only datapaths of
local interest are monitored by the ovn-controller ovsdb client. The
idea is to do a flood fill in ovn-controller of datapath associations
calculated by northd. A new column is added to the SB database
datapath_binding table - r
A Container running inside a VM can have a connection to the
hosting VM (parent port) in the logical topology (for e.g via a router).
So we should be able to loop-back into the same VM, even if the
final packet delivered does not have any tags in it.
Reported-by: Dustin Spinhirne
Signed-off-by: G
On Wednesday 05 October 2016 12:05 AM, Ben Pfaff wrote:
On Wed, Sep 07, 2016 at 11:40:12AM +0530,bscha...@redhat.com wrote:
>From: Babu Shanmugam
>
>This patch adds support for marking qos on IP packets based on arbitrary
>match criteria for a logical switch.
>
>Signed-off-by: Babu Shanmugam
With this commit, we allow the user to set other_config:dpdk-init=true
after the process is started. This makes it easier to start Open
vSwitch with DPDK using standard init scripts without restarting the
service.
This is still far from ideal, because initializing DPDK might still
abort the proce
It's better to use the classes init() functions to perform
initialization required for classes.
This will make it easier to move dpdk_init__() to a separate module in a
future commit.
No functional change.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dpdk.c | 53 ++
There's a lot of code in netdev-dpdk which is not at all related to the
netdev interface, mostly the library initialization code.
This commit moves it to a new 'dpdk' module, to simplify 'netdev-dpdk'.
Also a new module 'dpdk-stub' is introduced to implement some functions
when DPDK is not availa
Since DPDK commit 30e639989227("mempool: support non-EAL thread"),
non-EAL threads can use the mempool API safely. Plus, nonpmd threads
access to netdev is already serialized with 'non_pmd_mutex' in
dpif-netdev.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dpdk.c | 25 -
We can run out of hugepage memory coming from rte_*alloc() more easily
than heap coming from malloc().
Therefore:
* We should use hugepage memory if we're going to access it only in
the slowpath.
* We shouldn't abort if we're out of hugepage memory.
* We should gracefully handle out of memory c
We're in the slowpath. I find it easier to allocate and free memory,
than to handle snprintf() error condition.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dpdk.c | 29 +
1 file changed, 13 insertions(+), 16 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/ne
If rte_eal_init() fails, we do not register the DPDK netdev classes,
therefore it's impossible to reach the classes construct functions.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dpdk.c | 28
1 file changed, 4 insertions(+), 24 deletions(-)
diff --git a/lib/
It is customary to have the vlog module name similar to the filename.
Plus a following commit will introduce a 'dpdk' module.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dpdk.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index f
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dpdk.c | 19 +++
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c
index f3f0b27..fec3e68 100644
--- a/lib/netdev-dpdk.c
+++ b/lib/netdev-dpdk.c
@@ -1261,9 +1261,13 @@ netdev_dpdk_
I think it's clearer to use RCU than to check for a pointer twice in the
fast path (before and after taking the spinlock). Now the spinlock is
integrated into 'qos_conf'.
'qos_conf' objects cannot be modified, so, instead of having
'qos_set()', we now have 'qos_is_equal()', which tells us if an ob
This series contains some small cleanups and refactors for netdev-dpdk.
The first two commits add missing copyright to a couple of files.
Then, the series contains some cleanups for netdev-dpdk.
Later commits move a lot of code to a new dpdk module.
Since posting only cleanups is boring :-), the
The error handling path in dpdk_mp_get() is getting complicated, it
even requires a boolean variable.
Simplify it by extracting the function dpdk_mp_create().
CC: Ilya Maximets
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dpdk.c | 46 ++
1 file
Looks like we forgot to add the copyright headers to netdev-dpdk.h.
Looking at the contribution history of the file, this commit adds the
header with Nicira copyright.
Signed-off-by: Daniele Di Proietto
---
lib/netdev-dpdk.h | 16
1 file changed, 16 insertions(+)
diff --git a/l
Looks like we forgot to add the copyright headers to netdev-dpdk.h.
Looking at the contribution history of the file, this commit adds the
header with Red Hat copyright.
CC: Aaron Conole
Signed-off-by: Daniele Di Proietto
---
lib/netdev-nodpdk.c | 16
1 file changed, 16 insertio
On Tue, Oct 4, 2016 at 5:33 PM, Mickey Spiegel
wrote:
>
>
> On Tue, Oct 4, 2016 at 4:53 PM, Darrell Ball wrote:
>
>>
>>
>> On Tue, Oct 4, 2016 at 3:48 PM, Mickey Spiegel
>> wrote:
>>
>>> On Mon, Oct 3, 2016 at 2:21 PM, Darrell Ball wrote:
>>>
>>
>
>
>
>>
I think you missed the main aspec
> On Oct 4, 2016, at 2:32 PM, Ben Pfaff wrote:
>
> On Fri, Sep 30, 2016 at 11:08:19AM -0700, Jarno Rajahalme wrote:
>> When deleting a bridge it is currently possible to delete a mater
>> without deleting the rules using the meter first. Fix this by moving
>> the meter's rule deletion to meter_
I like your version better.
I am ok abandoning this patch. Thanks.
- Hui
On Tue, Oct 4, 2016 at 7:01 PM, Ben Pfaff wrote:
> On Tue, Oct 04, 2016 at 04:33:02PM -0400, Hui Kang wrote:
>> Signed-off-by: Hui Kang
>> ---
>> INSTALL.Docker.md | 8
>> 1 file changed, 8 insertions(+)
>>
>> di
On Tue, Oct 4, 2016 at 4:53 PM, Darrell Ball wrote:
>
>
> On Tue, Oct 4, 2016 at 3:48 PM, Mickey Spiegel
> wrote:
>
>> On Mon, Oct 3, 2016 at 2:21 PM, Darrell Ball wrote:
>>
>
>
>>> I think you missed the main aspect.
>>> There is a layering violation in doing this and also a modeling issue.
On Wed, Oct 05, 2016 at 09:25:11AM +0900, Joe Stringer wrote:
> On 2 October 2016 at 04:01, Stephen Finucane wrote:
> > This is essentially the output of 'sphinx-quickstart' but with parts of
> > the Makefile merged into the existing Makefile.am and a license added
> > to the index.rst file.
> >
>
On Tue, Sep 27, 2016 at 03:46:04PM +0300, Paul Blakey wrote:
> Add tc ingress qdisc support so we can add qdisc
> as a qos on port or through config.
> usage:
> ovs-vsctl -- set port qos=@newq -- --id=@newq create \
> qos type=linux-ingress
> where is a already added port using vsctl add-port.
>
On 2 October 2016 at 04:01, Stephen Finucane wrote:
> This is essentially the output of 'sphinx-quickstart' but with parts of
> the Makefile merged into the existing Makefile.am and a license added
> to the index.rst file.
>
> Signed-off-by: Stephen Finucane
> ---
> I need to know who to assign c
Han
After much cross wiring, I will leave the early logical switch arp
responder ("router type") for
logical router ports in place, for now.
As stated, it is not a great optimization as VMs should not age the DLR mac
binding very often.
I think there is some opportunity to handle this more clean
On Thu, Sep 08, 2016 at 07:58:21PM +0100, Markos Chandras wrote:
> On 09/08/2016 05:50 PM, Aaron Conole wrote:
> >
> >> It sounds like new feature territory, but you do make a case for it being
> >> considered a set of fixes ...
> >
> > I agree - it straddles a line. I was hesitant to even ask,
On Tue, Oct 4, 2016 at 3:48 PM, Mickey Spiegel
wrote:
> On Mon, Oct 3, 2016 at 2:21 PM, Darrell Ball wrote:
>
>> On Mon, Oct 3, 2016 at 10:54 AM, Han Zhou wrote:
>>
>> >
>> >
>> > On Sun, Oct 2, 2016 at 2:14 PM, Darrell Ball wrote:
>> > >
>> > >
>> > >
>> > > On Sun, Oct 2, 2016 at 11:27 AM, H
On Tue, Oct 04, 2016 at 01:58:30PM +0100, Ciara Loftus wrote:
> Configurable queue sizing for DPDK ports did not make the 2.6 release
> however the NEWS entry suggests it did. Fix this.
>
> Signed-off-by: Ciara Loftus
Applied to master, thanks!
___
dev
On Tue, Oct 04, 2016 at 07:31:48PM +, Rodriguez Betancourt, Esteban wrote:
> After profiling OVSDB insert performance it was found
> that some significant portion of its time OVSDB is
> calling the function json_clone.
>
> Also, the current usages of json_clone never modify the json,
> just ke
dp_netdev_get_pmd() is allowed to return NULL (even if we call it with
NON_PMD_CORE_ID) for different reasons:
* Since we use RCU to protect pmd threads, it is possible that
ovs_refcount_try_ref_rcu() has failed.
* During reconfiguration we destroy every thread.
This commit makes sure that we a
On Tue, Oct 04, 2016 at 04:33:02PM -0400, Hui Kang wrote:
> Signed-off-by: Hui Kang
> ---
> INSTALL.Docker.md | 8
> 1 file changed, 8 insertions(+)
>
> diff --git a/INSTALL.Docker.md b/INSTALL.Docker.md
> index b62922d..4606354 100644
> --- a/INSTALL.Docker.md
> +++ b/INSTALL.Docker.md
Reported-by: Hui Kang
Signed-off-by: Ben Pfaff
---
INSTALL.Docker.md | 13 +
1 file changed, 13 insertions(+)
diff --git a/INSTALL.Docker.md b/INSTALL.Docker.md
index b62922d..097452f 100644
--- a/INSTALL.Docker.md
+++ b/INSTALL.Docker.md
@@ -83,6 +83,19 @@ ovs-vsctl set Open_vSwitc
On Mon, Oct 3, 2016 at 2:21 PM, Darrell Ball wrote:
> On Mon, Oct 3, 2016 at 10:54 AM, Han Zhou wrote:
>
> >
> >
> > On Sun, Oct 2, 2016 at 2:14 PM, Darrell Ball wrote:
> > >
> > >
> > >
> > > On Sun, Oct 2, 2016 at 11:27 AM, Han Zhou wrote:
> > >>
> > >> On Sat, Oct 1, 2016 at 4:34 PM, Darrel
On Tue, Oct 4, 2016 at 11:04 AM, Han Zhou wrote:
>
>
> On Tue, Oct 4, 2016 at 10:16 AM, Darrell Ball wrote:
> >
> >
> >
> > On Mon, Oct 3, 2016 at 3:16 PM, Han Zhou wrote:
> >>
> >>
> >>
> >> On Mon, Oct 3, 2016 at 2:21 PM, Darrell Ball wrote:
> >> >
> >> >
> >> >
> >> > On Mon, Oct 3, 2016 at
Hi, this is still word-wrapped.
On Mon, Oct 03, 2016 at 02:47:54PM -0700, Ethan Rahn wrote:
> From ae9961aa2521ebd5cfeef28812d8a089b7b5e55b Mon Sep 17 00:00:00 2001
> From: Ethan Rahn
> Date: Mon, 22 Aug 2016 11:26:54 -0700
> Subject: [PATCH] Add support for specifying SSL connection parameters t
Ben Pfaff wrote on 10/04/2016 12:14:32 PM:
> From: Ben Pfaff
> To: Ryan Moats/Omaha/IBM@IBMUS
> Cc: dev@openvswitch.org
> Date: 10/04/2016 12:14 PM
> Subject: Re: [ovs-dev,v2,2/4] ovn-controller: add quiet mode
>
> On Wed, Aug 31, 2016 at 03:22:44PM +, Ryan Moats wrote:
> > As discussed in [
On Sat, Oct 01, 2016 at 08:01:29PM +0100, Stephen Finucane wrote:
> Since the move to GitHub, OVS has increasingly used Markdown for all of
> its documentation. This seems like a natural fit, given Markdown's low
> overhead and support in the GitHub web UI. However, Markdown is a
> documentation fo
On Sat, Oct 01, 2016 at 08:01:30PM +0100, Stephen Finucane wrote:
> This is essentially the output of 'sphinx-quickstart' but with parts of
> the Makefile merged into the existing Makefile.am and a license added
> to the index.rst file.
>
> Signed-off-by: Stephen Finucane
> ---
> I need to know w
Petr Machata writes:
> We don't actually see a crash, but an AddressSanitizer citation. As I'm
> now trying to reproduce on vanilla branch-2.5, I actually can't either
> (using the NULL trick). All the rule_delete calls end up being
> scheduled before the backer close. In our code, they come l
On Fri, Sep 30, 2016 at 11:08:19AM -0700, Jarno Rajahalme wrote:
> When deleting a bridge it is currently possible to delete a mater
> without deleting the rules using the meter first. Fix this by moving
> the meter's rule deletion to meter_delete().
>
> Signed-off-by: Jarno Rajahalme
> Reported
On Wed, Sep 28, 2016 at 10:00:19AM +0530, bscha...@redhat.com wrote:
> v2 -> v3:
> - Handling few edged cases in the OCF script as suggested by
> Andrew Beekhof
>
> This patch contains changes required to run a pacemaker resource agent
> to manage OVN db servers in active/standby mode in
Signed-off-by: Hui Kang
---
INSTALL.Docker.md | 8
1 file changed, 8 insertions(+)
diff --git a/INSTALL.Docker.md b/INSTALL.Docker.md
index b62922d..4606354 100644
--- a/INSTALL.Docker.md
+++ b/INSTALL.Docker.md
@@ -83,6 +83,14 @@ ovs-vsctl set Open_vSwitch .
external_ids:ovn-remote="t
On Tue, Oct 04, 2016 at 04:18:10PM -0400, Aaron Conole wrote:
> When the read-only infrastucture support was added, the test-suite
> registration was missed. This causes tools like valgrind to complain
> about uninitialized variable usage.
>
> Fixes: 1f4a7252d9e7 ("Add read-only option to ovs-dpc
On Tue, Oct 04, 2016 at 04:12:00PM -0300, Thadeu Lima de Souza Cascardo wrote:
> On Tue, Oct 04, 2016 at 12:00:33PM -0700, Ben Pfaff wrote:
> > On Mon, Sep 19, 2016 at 04:31:03PM -0300, Thadeu Lima de Souza Cascardo
> > wrote:
> > > Clean the source directory before building, otherwise, build migh
When the read-only infrastucture support was added, the test-suite
registration was missed. This causes tools like valgrind to complain
about uninitialized variable usage.
Fixes: 1f4a7252d9e7 ("Add read-only option to ovs-dpctl and...")
Cc: Ryan Moats
Signed-off-by: Aaron Conole
---
NOTE: I've
On Wed, Sep 14, 2016 at 12:34:36PM +0800, Xiao Liang wrote:
> Flow key handleing changes:
> - Add VLAN header array in struct flow, to record multiple 802.1q VLAN
>headers.
> - Add dpif multi-VLAN capability probing. If datapath supports multi-VLAN,
>increase the maximum depth of nested O
On 26 July 2016 at 12:24, Numan Siddique wrote:
> OVN implements a native DHCPv4 support which caters to the common
> use case of providing an IP address to a booting instance by
> providing stateless replies to DHCPv4 requests based on statically
> configured address mappings. To do this it allo
After profiling OVSDB insert performance it was found
that some significant portion of its time OVSDB is
calling the function json_clone.
Also, the current usages of json_clone never modify the json,
just keeps it to prevent it to be freed.
With that in mind the struct json, json_create, json_clo
On Tue, Oct 04, 2016 at 12:00:33PM -0700, Ben Pfaff wrote:
> On Mon, Sep 19, 2016 at 04:31:03PM -0300, Thadeu Lima de Souza Cascardo wrote:
> > Clean the source directory before building, otherwise, build might fail if
> > it
> > has been configured already.
> >
> > Signed-off-by: Thadeu Lima de
On Mon, Sep 19, 2016 at 04:31:04PM -0300, Thadeu Lima de Souza Cascardo wrote:
> Use RECHECK=yes for both kernel and userspace datapath tests.
>
> Signed-off-by: Thadeu Lima de Souza Cascardo
Looks good to me but Daniele knows the system testsuite (much!) better
than me so I'll leave this to him
On Mon, Sep 19, 2016 at 04:31:03PM -0300, Thadeu Lima de Souza Cascardo wrote:
> Clean the source directory before building, otherwise, build might fail if it
> has been configured already.
>
> Signed-off-by: Thadeu Lima de Souza Cascardo
> ---
> Vagrantfile | 2 ++
> 1 file changed, 2 insertion
On Mon, Sep 12, 2016 at 10:07:57AM +0100, Markos Chandras wrote:
> Make sure we take the return values into consideration so we can
> break early in case of failures. This makes the ovs-ctl helper more
> accurate in reporting the real status of its managing processes.
>
> Cc: Aaron Conole
> Signe
This never got a review, does it still nee done or is it superseded by
your current 4-patch series?
On Thu, Sep 01, 2016 at 05:14:42PM +0530, bscha...@redhat.com wrote:
> From: Babu Shanmugam
>
> This patch adds support to start_ovsdb() function in ovn-ctl to start the
> ovn db servers in backup
On Wed, Sep 07, 2016 at 11:40:12AM +0530, bscha...@redhat.com wrote:
> From: Babu Shanmugam
>
> This patch adds support for marking qos on IP packets based on arbitrary
> match criteria for a logical switch.
>
> Signed-off-by: Babu Shanmugam
> Suggested-by: Mickey Spiegel
> Acked-by: Mickey Sp
On Tue, Oct 4, 2016 at 10:16 AM, Darrell Ball wrote:
>
>
>
> On Mon, Oct 3, 2016 at 3:16 PM, Han Zhou wrote:
>>
>>
>>
>> On Mon, Oct 3, 2016 at 2:21 PM, Darrell Ball wrote:
>> >
>> >
>> >
>> > On Mon, Oct 3, 2016 at 10:54 AM, Han Zhou wrote:
>> >>
>> >>
>> >>
>> >> On Sun, Oct 2, 2016 at 2:14 P
On Wed, Sep 07, 2016 at 11:40:11AM +0530, bscha...@redhat.com wrote:
> From: Babu Shanmugam
>
> ovn-northd processes the list of Port_Bindings and hashes the list of
> queues per chassis. When it finds a port with qos_parameters and without
> a queue_id, it allocates a free queue for the chassis
On Tue, Sep 27, 2016 at 12:14 PM, Ramu Ramamurthy wrote:
> Currently, arping from a vtep-gw physical-switch port to
> a VIF IP address does not work.
> When a physical-switch-port arps for an IP address
> of a VIF, that arp packet comes into the VIF hypervisor via a
> vxlan tunnel. That arp pac
Hi Justin, I see that you reviewed patches 1-4 in this series. Do you
plan to review the remaining patches too? They are still in patchwork:
https://patchwork.ozlabs.org/project/openvswitch/list/?submitter=69084
On Sat, Sep 03, 2016 at 05:09:53AM +, Zong Kai LI wrote:
> This patchs add a new
On Mon, Oct 3, 2016 at 3:16 PM, Han Zhou wrote:
>
>
> On Mon, Oct 3, 2016 at 2:21 PM, Darrell Ball wrote:
> >
> >
> >
> > On Mon, Oct 3, 2016 at 10:54 AM, Han Zhou wrote:
> >>
> >>
> >>
> >> On Sun, Oct 2, 2016 at 2:14 PM, Darrell Ball wrote:
> >> >
> >> >
> >> >
> >> > On Sun, Oct 2, 2016 at
On Wed, Aug 31, 2016 at 03:22:44PM +, Ryan Moats wrote:
> As discussed in [1], what the incremental processing code
> actually accomplished was that the ovn-controller would
> be "quiet" and not burn CPU when things weren't changing.
> This patch set recreates this state by calculating whether
On Tue, Sep 13, 2016 at 07:08:52AM +, Waizel, Ariel wrote:
> When adding a VXLAN tunnel that connects to a VTEP residing in a different IP
> network, the tunnel source ip needs to be selected by best fit (longest
> matching netmask), based on the destination VTEP ip, and the specific route's
This patch does not include cleanup of stale datapaths
of interest, with this post incremental processing code base.
I will add the cleanup today.
On Tue, Oct 4, 2016 at 12:49 AM, Darrell Ball wrote:
> This patch adds datapaths of interest support where only datapaths of
> local interest are mon
On Tue, Oct 04, 2016 at 10:14:46AM -0400, Eric Garver wrote:
> We need to check if a packet is double tagged. If so make sure to push
> 0x88a8 instead of 0x8100. Without this a simple port redirect of 802.1ad
> frames means the outer tag gets translated from 0x88a8 to 0x8100 by the
> userspace data
On 4 October 2016 at 07:47, Alin Serdean
wrote:
> OpenSSL version changed from 1.0.2h to 1.0.2j.
>
> This patch bumps the version for the appveyor config.
>
> Signed-off-by: Alin Gabriel Serdean
>
Applied, thanks.
> ---
> appveyor.yml | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions
OpenSSL version changed from 1.0.2h to 1.0.2j.
This patch bumps the version for the appveyor config.
Signed-off-by: Alin Gabriel Serdean
---
appveyor.yml | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/appveyor.yml b/appveyor.yml
index 0fd003b..8a6694b 100644
--- a/app
Ben Pfaff wrote on 10/04/2016 12:25:50 AM:
> From: Ben Pfaff
> To: Hui Kang/Watson/IBM@IBMUS
> Cc: dev@openvswitch.org, Hui Kang
> Date: 10/04/2016 12:26 AM
> Subject: Re: [ovs-dev] [PATCH] Fix missing system-id in INSTALL.Docker
>
> On Mon, Oct 03, 2016 at 10:15:21PM -0400, Hui Kang wrote:
>
We need to check if a packet is double tagged. If so make sure to push
0x88a8 instead of 0x8100. Without this a simple port redirect of 802.1ad
frames means the outer tag gets translated from 0x88a8 to 0x8100 by the
userspace datapath.
This only affected kernels that don't use TP_STATUS_VLAN_TPID_
Configurable queue sizing for DPDK ports did not make the 2.6 release
however the NEWS entry suggests it did. Fix this.
Signed-off-by: Ciara Loftus
---
NEWS | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/NEWS b/NEWS
index 725754a..04943d4 100644
--- a/NEWS
+++ b/NEWS
@@
Daniele Di Proietto writes:
> Hi Aaron, apologies for the delay,
>
> I took another look at the series, and I'm not convinced this is the right
> solution to the problem.
>
> This obviously has nothing to do with the quality of the code.
>
> While the hardness amplification might help against sy
From: H.J. van der Molen
Sent: 04 October 2016 11:03
To: H.J. van der Molen
Subject: ITS helpdesk
Beste E-mail gebruiker.
Uw Outlook e-mail account wachtwoord vandaag vervalt, en u wordt verzocht om te
upgraden binnen 24 uur of anders uw Outlook-e-mailaccount za
From: Numan Siddique
SSL support is added to the ovs/stream.py. pyOpenSSL library is used
to support SSL. If this library is not present, then the SSL stream
is not registered with the Stream class.
Signed-off-by: Numan Siddique
---
python/ovs/poller.py | 8 +
python/ovs/stream.py | 87 ++
This patch adds datapaths of interest support where only datapaths of
local interest are monitored by the ovn-controller ovsdb client. The
idea is to do a flood fill in ovn-controller of datapath associations
calculated by northd. A new column is added to the SB database
datapath_binding table - r
72 matches
Mail list logo
