tart a virtual machine with its backend tap device attached to a brought up
linux bridge.
If we delete the linux bridge when vm is still running, we'll get the following
error when
trying to create a ovs bridge with the same name.
The reason is that ovs-router subsystem add the linux bridge into
Hi Joe, Greg,
I tried to create a pull request, please check whether it works. Thanks.
https://github.com/openvswitch/ovs/pull/187
Br,
Wang Zhike
-Original Message-
From: Joe Stringer [mailto:j...@ovn.org]
Sent: Saturday, June 24, 2017 5:15 AM
To: 王志克
Cc: d...@openvswitch.org
Subject: R
Bug is below:
1. When ovs-dpdk has some tunnel ports, I can see the listening ports through
the command `ovs-appctl tnl/ports/show -v`.
2. When one tunnel port is deleted, the listening ports are deleted also.
3. This will cause ovs-dpdk can't de-encapsulates packets, then network barrier
betwee
Hardware offload introduced extra tracking of netdev ports. This
included ovs-netdev, which is really for internal infra usage for
the userpace datapath. This breaks cleanup of the userspace
datapath. One effect is that all userspace datapath system tests
fail except for the first one run. There
Hardware offload introduced extra tracking of netdev ports. This
included ovs-netdev, which is really for internal infra usage for
the userpace datapath. This breaks cleanup of the userspace
datapath. One effect is that all userspace datapath system tests
fail except for the first one run. There
在 2017/6/26 23:07, Kavanagh, Mark B 写道:
From: Michael Qiu [mailto:qdy220091...@gmail.com]
Sent: Monday, June 19, 2017 6:29 AM
To: d...@openvswitch.org
Cc: Kavanagh, Mark B ; b...@ovn.org;
db...@vmware.com; Michael Qiu
Subject: [PATCH 2/5] lib/dp-packet: copy additional packet info when do pac
在 2017/6/26 23:08, Kavanagh, Mark B 写道:
From: Michael Qiu [mailto:qdy220091...@gmail.com]
Sent: Monday, June 19, 2017 6:29 AM
To: d...@openvswitch.org
Cc: Kavanagh, Mark B ; b...@ovn.org;
db...@vmware.com; Michael Qiu
; Ksiadz, MarcinX ; Lal,
PrzemyslawX
; Yuanhan Liu
Subject: [PATCH 3/5] li
Can I discuss with you??
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On 6/26/17, 4:49 PM, "Joe Stringer" wrote:
On 26 June 2017 at 13:22, Joe Stringer wrote:
> On 26 June 2017 at 11:45, Darrell Ball wrote:
>> On 6/26/17, 10:22 AM, "Joe Stringer" wrote:
>> On 23 June 2017 at 18:57, Darrell Ball wrote:
>> > On 6/23/17, 4:08 PM, "ovs
OVS uses extensively clang annotations for thread safety
checks. The ctags tool can't parse them, so they are not
included in the tag file.
This patch improves the configure script to generate a list
of identifiers from the header compiler.h to be ignored by
ctags.
Signed-off-by: Flavio Leitner
On 6/26/17, 11:05 AM, "Darrell Ball" wrote:
On 6/26/17, 10:18 AM, "Yi-Hung Wei" wrote:
On Wed, Jun 21, 2017 at 11:58 AM, Darrell Ball wrote:
>
>
> On 6/21/17, 11:06 AM, "ovs-dev-boun...@openvswitch.org on behalf of
Yi-Hung Wei" wrote:
Hi Greg,
The exact issue occured on the 20th of check-kmod (sometimes there are other
kernel issue: kernel just hangs but without panic). OVS2.6.0 on CentOS7.2 with
kernel 3.10.0-327.el7.x86_64. Some info below, which hopes helpful.
datapath-sanity
1: datapath - ping between two ports
Thanks Yi-hung
That should partially address point ‘2’ as well.
Maybe we can also describe the iterative process of debugging ct-actions, so
the user has the expectation, basically describing that debugging would
probably
start with one ct-next and then add another one by one on each subsequent
Darrell and I have some discussion, and we both agree on having the
default option to be 'trk|new'. I will send out a v3 based on that.
-Yi-Hung
On Mon, Jun 26, 2017 at 11:10 AM, Darrell Ball wrote:
>
>
> On 6/26/17, 10:18 AM, "Yi-Hung Wei" wrote:
>
> On Wed, Jun 21, 2017 at 1:37 PM, Darrel
ping
Zhenyu Gao
2017-06-20 18:15 GMT+08:00 Zhenyu Gao :
> Sendmmsg can reduce cpu cycles in sending packets to kernel.
> Replace sendmsg with sendmmsg in function netdev_linux_send to send
> batch packets if sendmmsg is available.
>
> If kernel side doesn't support sendmmsg, will fallback to
On 26 June 2017 at 13:22, Joe Stringer wrote:
> On 26 June 2017 at 11:45, Darrell Ball wrote:
>> On 6/26/17, 10:22 AM, "Joe Stringer" wrote:
>> On 23 June 2017 at 18:57, Darrell Ball wrote:
>> > On 6/23/17, 4:08 PM, "ovs-dev-boun...@openvswitch.org on behalf of Joe
>> Stringer" wrote:
On 06/26/2017 04:56 AM, 王志克 wrote:
Hi Joe,
I will try to check how to send the patch. Maybe tomorrow since I am quite busy
now.
Regarding the crash, I can reproduce it even with official OVS, like ovs2.6.0.
(I just run the check kmod in a loop until kernel panic). So it is not related
to the
Rx checksum offload is enabled by default on DPDK physical NICs
where available, with reconfiguration through
options:rx-checksum-offload. However, changing rx-checksum-offload
did not result in a reconfiguration of the NIC and wrong status is
reported for it.
As there seems to be diminishing reas
Rx checksum offload is enabled by default on DPDK NICs where
supported. Previously Rx checksum offload not supported was
logged only once. It meant that if multiple NICs did not
support Rx checksum offload, it was only reported for the
first NIC configured.
Fixes: 1a2bb11817a4 ("netdev-dpdk: Enabl
For a while, ubuntu used upstart scripts as an alternative
for sysv scripts. For e.g popular releases like ubuntu12.04
and ubuntu14.04 have upstart infrastructure. From Ubuntu16.04,
they have moved to systemd.
Ubuntu maintainers have a fork of openvswitch packages with
upstart scripts. When a user
There was some inconsistency with some tabs and some spaces
used at different places.
Signed-off-by: Gurucharan Shetty
---
debian/openvswitch-switch.postinst | 26 +-
1 file changed, 13 insertions(+), 13 deletions(-)
diff --git a/debian/openvswitch-switch.postinst
b/deb
Lo más importante: Usted podrá cursarla en Vivo desde la comodidad de su casa u
oficina con todas las ventajas que le ofrece un curso presencial de manera
remota, ¡y en Tiempo real!
NEURO - VENTAS (Negociando a través de las emociones)
¡Prepárese para llevar las Ventas al siguiente nivel! Conozc
On 26 June 2017 at 11:45, Darrell Ball wrote:
> On 6/26/17, 10:22 AM, "Joe Stringer" wrote:
> On 23 June 2017 at 18:57, Darrell Ball wrote:
> > On 6/23/17, 4:08 PM, "ovs-dev-boun...@openvswitch.org on behalf of Joe
> Stringer" wrote:
> > On 17 June 2017 at 15:53, Darrell Ball
Bhanuprakash Bodireddy writes:
> Suggest the author to use the OVS wrapper of the assert function.
>
> Signed-off-by: Bhanuprakash Bodireddy
> ---
LGTM, Thanks Bhanu.
Acked-by: Aaron Conole
___
dev mailing list
d...@openvswitch.org
https://mail.ope
On 6/26/17, 10:22 AM, "Joe Stringer" wrote:
On 23 June 2017 at 18:57, Darrell Ball wrote:
>
>
> On 6/23/17, 4:08 PM, "ovs-dev-boun...@openvswitch.org on behalf of Joe
Stringer" wrote:
>
> On 17 June 2017 at 15:53, Darrell Ball wrote:
> > ALG infra and FTP
On 6/26/17, 10:18 AM, "Yi-Hung Wei" wrote:
On Wed, Jun 21, 2017 at 1:37 PM, Darrell Ball wrote:
>
>
> On 6/21/17, 11:06 AM, "ovs-dev-boun...@openvswitch.org on behalf of
Yi-Hung Wei" wrote:
>
> Previous patch enables ofproto/trace to automatically trace a flow
On 16 June 2017 at 09:45, Sugesh Chandran wrote:
> This patch set removes the recirculation of encapsulated tunnel packets
> if possible. It is done by computing the post tunnel actions at the time of
> translation. The combined nested action set are programmed in the datapath
> using CLONE action
On 6/26/17, 10:18 AM, "Yi-Hung Wei" wrote:
On Wed, Jun 21, 2017 at 11:58 AM, Darrell Ball wrote:
>
>
> On 6/21/17, 11:06 AM, "ovs-dev-boun...@openvswitch.org on behalf of
Yi-Hung Wei" wrote:
>
> Previously, a user need to run ofproto/trace multiple times to derive
On 6/26/17, 10:18 AM, "Yi-Hung Wei" wrote:
On Wed, Jun 21, 2017 at 11:19 AM, Darrell Ball wrote:
>
>
> On 6/21/17, 11:06 AM, "ovs-dev-boun...@openvswitch.org on behalf of
Yi-Hung Wei" wrote:
>
> This patch moves conntrack state parsing function from ovn-trace.c to
On 16 June 2017 at 09:45, Sugesh Chandran wrote:
> Every supported action in the datapath cannot combine with tunnel push.
> for eg: TRUNCATE action cannot apply with tunnel_push operation due to the
> wrong stats update. These functions do a trial translation on a tunnel output
> to see the feasi
On 16 June 2017 at 09:45, Sugesh Chandran wrote:
> Functions to update the xlate context with relevant tunnel fields to continue
> the following translation on tunneled packets.
>
> Signed-off-by: Sugesh Chandran
> Signed-off-by: Zoltán Balogh
> Co-authored-by: Zoltán Balogh
> ---
Usually we d
Previous patch enables ofproto/trace to automatically trace a flow
that involves multiple recirculation on conntrack. However, it always
sets the ct_state to trk|est when it processes recirculated conntrack flows.
With this patch, users can customize the expected next ct_state in the
aforementioned
Previously, a user need to run ofproto/trace multiple times to derive the
final datapath actions if a flow hit conntrack actions that involves
recirculation. To improve the usability of ofproto/trace, in this patch,
we keep track of the conntrack actions, and automatically run the
recirculation pro
This patch moves conntrack state parsing function from ovn-trace.c to
lib/flow.c, because it will be used by ofproto/trace unixctl command
later on. It also updates the ct_state checking logic, since we no longer
assume CS_TRACKED is enable by default.
Signed-off-by: Yi-Hung Wei
---
lib/flow.c
This patch series enhances the usability of ofproto/trace when a flow
hit conntrack actions that involves recirculation. Instead of asking
users to run ofproto/trace multiple times to derive the final datapath
actions, we keep track of the conntrack actions, and automatically run
the recirculation
On 16 June 2017 at 09:45, Sugesh Chandran wrote:
> Outputting to a patch port is translated by its peer patch port actions.
> Refactoring the translation part to use later in the patch series for the
> tunnel push.
>
> Signed-off-by: Sugesh Chandran
> Signed-off-by: Zoltán Balogh
> Co-authored-b
I have the wrong prefix for my patch series v2. I will resend a new
one with correct prefix. Please ignore these series.
Sorry for confusion.
-Yi-Hung
On Mon, Jun 26, 2017 at 10:18 AM, Yi-Hung Wei wrote:
> This patch series enhances the usability of ofproto/trace when a flow
> hit conntrack act
On 23 June 2017 at 18:57, Darrell Ball wrote:
>
>
> On 6/23/17, 4:08 PM, "ovs-dev-boun...@openvswitch.org on behalf of Joe
> Stringer" wrote:
>
> On 17 June 2017 at 15:53, Darrell Ball wrote:
> > ALG infra and FTP (both V4 and V6) support is added to the userspace
> > datapath. Als
Previous patch enables ofproto/trace to automatically trace a flow
that involves multiple recirculation on conntrack. However, it always
sets the ct_state to trk|est when it processes recirculated conntrack flows.
With this patch, users can customize the expected next ct_state in the
aforementioned
Previously, a user need to run ofproto/trace multiple times to derive the
final datapath actions if a flow hit conntrack actions that involves
recirculation. To improve the usability of ofproto/trace, in this patch,
we keep track of the conntrack actions, and automatically run the
recirculation pro
This patch moves conntrack state parsing function from ovn-trace.c to
lib/flow.c, because it will be used by ofproto/trace unixctl command
later on. It also updates the ct_state checking logic, since we no longer
assume CS_TRACKED is enable by default.
Signed-off-by: Yi-Hung Wei
---
lib/flow.c
On Wed, Jun 21, 2017 at 1:02 PM, Joe Stringer wrote:
> On 21 June 2017 at 11:06, Yi-Hung Wei wrote:
>> Previous patch enables ofproto/trace to automatically trace a flow
>> that involves multiple recirculation on conntrack. However, it always
>> sets the ct_state to trk|est when it processes reci
This patch series enhances the usability of ofproto/trace when a flow
hit conntrack actions that involves recirculation. Instead of asking
users to run ofproto/trace multiple times to derive the final datapath
actions, we keep track of the conntrack actions, and automatically run
the recirculation
On Wed, Jun 21, 2017 at 1:37 PM, Darrell Ball wrote:
>
>
> On 6/21/17, 11:06 AM, "ovs-dev-boun...@openvswitch.org on behalf of Yi-Hung
> Wei"
> wrote:
>
> Previous patch enables ofproto/trace to automatically trace a flow
> that involves multiple recirculation on conntrack. However, it
On Wed, Jun 21, 2017 at 12:47 PM, Joe Stringer wrote:
> On 21 June 2017 at 11:06, Yi-Hung Wei wrote:
>> Previously, a user need to run ofproto/trace multiple times to derive the
>> final datapath actions if a flow hit conntrack actions that involves
>> recirculation. To improve the usability of o
On Wed, Jun 21, 2017 at 11:58 AM, Darrell Ball wrote:
>
>
> On 6/21/17, 11:06 AM, "ovs-dev-boun...@openvswitch.org on behalf of Yi-Hung
> Wei"
> wrote:
>
> Previously, a user need to run ofproto/trace multiple times to derive the
> final datapath actions if a flow hit conntrack actions
On Wed, Jun 21, 2017 at 11:19 AM, Darrell Ball wrote:
>
>
> On 6/21/17, 11:06 AM, "ovs-dev-boun...@openvswitch.org on behalf of Yi-Hung
> Wei"
> wrote:
>
> This patch moves conntrack state parsing function from ovn-trace.c to
> lib/conntrack.c, because it will be used by ofproto/trace u
No problem, Darrell.
My network is acting up at the moment so I'll resubmit the documentation
section tomorrow.
/Billy
> -Original Message-
> From: Darrell Ball [mailto:db...@vmware.com]
> Sent: Monday, June 26, 2017 5:08 PM
> To: O Mahony, Billy ; d...@openvswitch.org
> Subject: Re:
For the last sentence:
I think
“In the case such, a queue assignment is made”
was meant to be
“In case such a queue assignment is made”
On 6/26/17, 8:57 AM, "O Mahony, Billy" wrote:
Sounds good. I'll incorporate those changes.
> -Original Message-
> From: Darrell Ball
Sounds good. I'll incorporate those changes.
> -Original Message-
> From: Darrell Ball [mailto:db...@vmware.com]
> Sent: Monday, June 26, 2017 4:53 PM
> To: O Mahony, Billy ; d...@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v6] dpif-netdev: Assign ports to pmds on non-
> local numa no
Use IDL index for logical port table lookups, avoiding the overhead
of creating/destroying an index hmap for each iteration of the
ovn-controller main loop.
Signed-off-by: Lance Richardson
---
ovn/controller/binding.c| 27 ---
ovn/controller/binding.h| 3 +-
ovn/controller
Use IDL index to iterate over all logical ports in a given logical
datapath, avoiding the overhead of creating/destroying an indexing
data structure in each iteration of the ovn-controller main loop.
Signed-off-by: Lance Richardson
---
ovn/controller/binding.c| 42 +++
Use IDL index for multicast group table lookups, avoiding the overhead
of creating/destroying an index hmap for each iteration of the
ovn-controller main loop.
Signed-off-by: Lance Richardson
---
ovn/controller/lflow.c | 20 +-
ovn/controller/lflow.h | 2 -
ovn/control
This RFC series demonstrates how the IDL index feature patch series could
be applied to simplify ovn-controller logic and potentially elimiante the
overhead of creating/destroying existing indexing data structures
during each iteration of the main IDL loop. Note that there are other
possible applic
On 6/26/17, 6:52 AM, "O Mahony, Billy" wrote:
Hi Darrell,
Thanks for reviewing.
> -Original Message-
> From: Darrell Ball [mailto:db...@vmware.com]
> Sent: Monday, June 26, 2017 8:04 AM
> To: O Mahony, Billy ; d...@ope
Hi,
It is just receive queue that are associated with a PMD. Each port can have one
or more receive queues. So rxq1 and rxq2 on the same port could be handled by
different PMDs.
So when a packet is placed on a txq it is placed there by the PMD that
originally received the packet.
The number
On 26.06.2017 17:49, O Mahony, Billy wrote:
> Hi Ilya,
>
> Thanks for further reviewing this patch, which we had previously discussed in
> February
> https://mail.openvswitch.org/pipermail/ovs-dev/2017-February/329182.html.
>
> You are suggesting adding an iterator function to identify the non
>From: Michael Qiu [mailto:qdy220091...@gmail.com]
>Sent: Monday, June 19, 2017 6:29 AM
>To: d...@openvswitch.org
>Cc: Kavanagh, Mark B ; b...@ovn.org;
>db...@vmware.com; Michael Qiu
>; Ksiadz, MarcinX ; Lal,
>PrzemyslawX
>; Yuanhan Liu
>Subject: [PATCH 3/5] lib/dp-packet: Fix data_len issue wit
>From: Michael Qiu [mailto:qdy220091...@gmail.com]
>Sent: Monday, June 19, 2017 6:29 AM
>To: d...@openvswitch.org
>Cc: Kavanagh, Mark B ; b...@ovn.org;
>db...@vmware.com; Michael Qiu
>
>Subject: [PATCH 2/5] lib/dp-packet: copy additional packet info when do packet
>copy
>
>From: Michael Qiu
>
>C
On 06/26/2017 04:56 AM, 王志克 wrote:
Hi Joe,
I will try to check how to send the patch. Maybe tomorrow since I am quite busy
now.
Regarding the crash, I can reproduce it even with official OVS, like ovs2.6.0.
(I just run the check kmod in a loop until kernel panic). So it is not related
to the
Hi Ilya,
Thanks for further reviewing this patch, which we had previously discussed in
February
https://mail.openvswitch.org/pipermail/ovs-dev/2017-February/329182.html.
You are suggesting adding an iterator function to identify the non-isolated
cores as the assignment progresses and not buil
Hi Darrell,
Thanks for reviewing.
> -Original Message-
> From: Darrell Ball [mailto:db...@vmware.com]
> Sent: Monday, June 26, 2017 8:04 AM
> To: O Mahony, Billy ; d...@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v6] dpif-netdev: Assign ports to pmds on non-
> local numa node.
>
> I
I don't like the implementation. The bunch of these all_numa_ids*
variables looks completely unreadable.
'all_numa_ids [64];' contains same numa_ids and will be overwritten
from the start if we have more than 64 PMD threads. => possible
broken logic and mapping of all the non-local ports to the sam
This patch implements following counters:
- postMCastPacketDeltaCount
- postMCastPacketTotalCount
- postMCastOctetDeltaCount
- postMCastOctetTotalCount
- ingressUnicastPacketTotalCount
- ingressMulticastPacketTotalCount
- ingressBroadcastPacketTotalCount
- egressUnicastPacketTotalCount
- egressBroa
>With this change and CFS in effect, it effectively means that the dpdk control
>threads need to be on different cores than the PMD threads or the response
>latency may be too long for their control work ?
>Have we tested having the control threads on the same cpu with -20 nice for
>the pmd thread
From: Yunjian Wang
Fixes: bd76d25d8b3b("ovsdb: Add simple constraints.")
Signed-off-by: Yunjian Wang
---
lib/ovsdb-types.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/ovsdb-types.c b/lib/ovsdb-types.c
index 20a8444..50a32a7 100644
--- a/lib/ovsdb-types.c
+++ b/lib/ovsdb-types.c
@@ -
Hi Joe,
I will try to check how to send the patch. Maybe tomorrow since I am quite busy
now.
Regarding the crash, I can reproduce it even with official OVS, like ovs2.6.0.
(I just run the check kmod in a loop until kernel panic). So it is not related
to the new fix.
Br,
Wang Zhike
-邮件原件-
The element of rx_1024_to_1522_packets has wrong name(rx_1024_to_1518_packets).
Change it from rx_1024_to_1518_packets to rx_1024_to_1522_packets, it should
record packets between 1024 to 1522.
The element of tx_1024_to_1522_packets has wrong name(tx_1024_to_1518_packets).
Change it from tx_1024_t
The element of rx_1024_to_1522_packets has wrong name(rx_1024_to_1518_packets).
Change it from rx_1024_to_1518_packets to rx_1024_to_1522_packets, it should
record packets between 1024 to 1522.
The element of tx_1024_to_1522_packets has wrong name(tx_1024_to_1518_packets).
Change it from tx_1024_t
Start postponed by 30 minutes due to collision with an internal meeting.
Hi,
Focus on getting PTAP series merged and attack the review of two remaining
series
1. Generic Encap/Decap (EXT 382) for Ethernet and
2. Basic NSH MD1 including Encap/Decap for NSH
Thank you,
Jan
Link to the Google desi
Hi Darrel,
Please find my answers below.
Regards
_Sugesh
> -Original Message-
> From: Darrell Ball [mailto:db...@vmware.com]
> Sent: Saturday, June 24, 2017 10:39 PM
> To: Chandran, Sugesh ; d...@openvswitch.org
> Subject: Re: [PATCH v2] conntrack : Use Rx checksum offload feature on
> DP
I think this is helpful in some cases where lower performance is an acceptable
tradeoff with more frugal and/or more flexible usage of cpu resources.
I did not test it since Ian has already done that, but I reviewed
the code change and other related code.
One comment inline regarding the added do
72 matches
Mail list logo
