[ovs-dev] Use more than 200000 flows in datapath/kernel cache flows

Dear ovs-dev, I'm trying to see if I can increase the number of flows supported in kernel cache. I found two relevant config options (other_config:flow-limit and upcall/flow limit). My understanding was that other_config:flow-limit is for the normal openflow flows while upcall is for the kernel

[ovs-dev] Herzlichen Glückwunsch und bitte antworten Sie uns jetzt

Ich bestätige den Erhalt der E-Mail. Ich bin eine Frau von Mavis Wanczyk, einer älteren Bürgerin der Vereinigten Staaten von Amerika und gebürtigen Massachusetts. Ich bin mit 14 Jahren aus Springfield gezogen und lebe jetzt in Chicopee. Am 25. August gewann ich die Powerball Lotterie im Wert

Re: [ovs-dev] [PATCHv6] netdev-afxdp: Add need_wakeup supprt.

On Mon, Oct 28, 2019 at 12:11 PM Ilya Maximets wrote: > > On 28.10.2019 11:46, Eelco Chaudron wrote: > > > > > > On 23 Oct 2019, at 23:06, William Tu wrote: > > > >> The patch adds support for using need_wakeup flag in AF_XDP rings. > >> A new option, use_need_wakeup, is added. When this option

Re: [ovs-dev] [PATCH v2.12] OVN: Combine conjunctions with identical matches into one flow.

Bleep bloop. Greetings Mark Michelson, I am a robot and I have tried out your patch. Thanks for your contribution. I encountered some error that I wasn't expecting. See the details below. git-am: Failed to merge in the changes. Patch failed at 0001 OVN: Combine conjunctions with identical

[ovs-dev] [PATCH v2.12] OVN: Combine conjunctions with identical matches into one flow.

This is a backport of commit e659bab31a916d540411c93ca7125011b2e82b5c from OVN master. Conjunctive matches have an issue where it is possible to install multiple flows that have identical matches. This results in ambiguity, and can lead to features (such as ACLs) not functioning properly. This

[ovs-dev] [PATCH ] travis: support ppc64le builds

Add support for travis-ci ppc64le builds. - Updated matrix in .travis.yml to include a ppc64le build. - Added support to install packages needed by specific architectures. To keep the total build time at an acceptable level only a single build job is included in the matrix for ppc64le. A build

Re: [ovs-dev] [PATCH 2/3] Combine conjunctions with identical matches into one flow.

I pushed the first two patches of the series to master. I'll resubmit patch three as a separate submission. On 10/28/19 1:45 PM, Numan Siddique wrote: On Mon, Oct 28, 2019, 9:54 PM Numan Siddique > wrote: On Mon, Oct 28, 2019, 9:29 PM Mark Michelson

Re: [ovs-dev] [PATCH ovs v3 2/2] netdev-dpdk: Add dpdkvdpa port

Hi Noa, Thanks for your reply. > > > > Hi Noa, > > > > > > > > I have a couple more questions. I'm still at the learning stage of > > > > this new feature, thanks in advance for your patience. > > > > > > > > On Thu, Oct 17, 2019 at 02:16:56PM +0300, Noa Ezra wrote: > > > > > dpdkvdpa netdev

Re: [ovs-dev] [PATCH ovs v3 0/2] Introduce dpdkvdpa netdev

Hi ilya, please see inline >-Original Message- >From: Ilya Maximets >Sent: Monday, October 28, 2019 3:46 PM >To: Noa Levy ; ovs-dev@openvswitch.org; Roni Bar Yanai > >Cc: Oz Shlomo ; Majd Dibbiny ; >Ameer Mahagneh ; Eli Britstein >; William Tu ; Simon Horman > >Subject: Re: [ovs-dev]

Re: [ovs-dev] [PATCHv6] netdev-afxdp: Add need_wakeup supprt.

On 28.10.2019 11:46, Eelco Chaudron wrote: On 23 Oct 2019, at 23:06, William Tu wrote: The patch adds support for using need_wakeup flag in AF_XDP rings. A new option, use_need_wakeup, is added.  When this option is used, it means that OVS has to explicitly wake up the kernel RX, using

Re: [ovs-dev] [PATCH branch-2.5] travis: Drop 2.6.32 kernel build.

On 06.08.2019 15:06, Ilya Maximets wrote: gcc >= 5 can't build Linux kernel 2.6.32 and this will never change because 2.6.32 is not supported for a last few years. TravsCI migrated to use Ubuntu Xenial by defualt with gcc 5 installed. Dropping the 2.6.32 build item to unlock green build.

Re: [ovs-dev] [PATCH ovn 00/19] OVN Interconnection

On Mon, Oct 28, 2019 at 7:25 AM Numan Siddique wrote: > > On Mon, Oct 21, 2019 at 6:23 AM Han Zhou wrote: > > > > The series supports interconnecting multiple OVN deployments (e.g. located at > > multiple data centers) through logical routers connected with tansit logical > > switches with

Re: [ovs-dev] Travis build failures due to base image change to Xenial.

On 28.10.2019 18:30, Ben Pfaff wrote: On Mon, Oct 28, 2019 at 06:24:44PM +0100, Ilya Maximets wrote: On 28.10.2019 18:05, Ben Pfaff wrote: On Mon, Oct 28, 2019 at 03:04:56PM +0100, Ilya Maximets wrote: On 09.08.2019 12:23, Ilya Maximets wrote: On 01.08.2019 20:48, Aaron Conole wrote: Ilya

Re: [ovs-dev] [PATCH 2/3] Combine conjunctions with identical matches into one flow.

On Mon, Oct 28, 2019, 9:54 PM Numan Siddique wrote: > > > On Mon, Oct 28, 2019, 9:29 PM Mark Michelson wrote: > >> On 10/28/19 11:33 AM, Numan Siddique wrote: >> > On Sat, Oct 26, 2019 at 2:37 AM Mark Michelson >> wrote: >> >> >> >> As stated in previous commits, conjunctive matches have an

Re: [ovs-dev] [PATCH 1/3] Revert conjunctive match removal patches.

On Sat, Oct 26, 2019, 2:38 AM Mark Michelson wrote: > This partially reverts commits 6f914327a55aaab11507db0911b08d695e17ce2a > and 298701dbc99645700be41680a43d049cb061847a. This restores the behavior > of making conjunctive matches, and it restores the tests to their state. > > The one thing it

Re: [ovs-dev] Travis build failures due to base image change to Xenial.

On Mon, Oct 28, 2019 at 06:24:44PM +0100, Ilya Maximets wrote: > On 28.10.2019 18:05, Ben Pfaff wrote: > > On Mon, Oct 28, 2019 at 03:04:56PM +0100, Ilya Maximets wrote: > > > On 09.08.2019 12:23, Ilya Maximets wrote: > > > > On 01.08.2019 20:48, Aaron Conole wrote: > > > > > Ilya Maximets

Re: [ovs-dev] [PATCH ovn] ovndb-servers.ocf: Change from 'openvswitch' to 'ovn' in default vars.

On Mon, Oct 28, 2019 at 10:25:34AM -0700, Ben Pfaff wrote: > On Mon, Oct 28, 2019 at 08:04:15PM +0530, num...@ovn.org wrote: > > From: Numan Siddique > > > > CC: Aliasgar Ginwala > > Signed-off-by: Numan Siddique > > Should we introduce something into the distro packaging, at the same > time,

Re: [ovs-dev] [PATCH ovn] ovndb-servers.ocf: Change from 'openvswitch' to 'ovn' in default vars.

On Mon, Oct 28, 2019 at 08:04:15PM +0530, num...@ovn.org wrote: > From: Numan Siddique > > CC: Aliasgar Ginwala > Signed-off-by: Numan Siddique Should we introduce something into the distro packaging, at the same time, to migrate from the old to the new locations automatically?

Re: [ovs-dev] Travis build failures due to base image change to Xenial.

On 28.10.2019 18:05, Ben Pfaff wrote: On Mon, Oct 28, 2019 at 03:04:56PM +0100, Ilya Maximets wrote: On 09.08.2019 12:23, Ilya Maximets wrote: On 01.08.2019 20:48, Aaron Conole wrote: Ilya Maximets writes: However, there is an additional issue with branch-2.5: branch-2.5 has kernel 2.6.32

Re: [ovs-dev] Travis build failures due to base image change to Xenial.

On Mon, Oct 28, 2019 at 03:04:56PM +0100, Ilya Maximets wrote: > On 09.08.2019 12:23, Ilya Maximets wrote: > > On 01.08.2019 20:48, Aaron Conole wrote: > > > Ilya Maximets writes: > > > > However, there is an additional issue with branch-2.5: > > > > > > > > branch-2.5 has kernel 2.6.32 in the

Re: [ovs-dev] [PATCH ovn] ovndb-servers.ocf: Change from 'openvswitch' to 'ovn' in default vars.

Thanks Numan. Acked-by: Aliasgar Ginwala mailto:num...@ovn.org>> From: num...@ovn.org Sent: Monday, October 28, 2019 7:34 AM To: d...@openvswitch.org Cc: Numan Siddique ; Ginwala, Aliasgar Subject: [PATCH ovn] ovndb-servers.ocf: Change from 'openvswitch' to

Re: [ovs-dev] [PATCH] dpif-netdev: Fix time delta overflow in case of race for meter lock.

On 25.10.2019 17:55, William Tu wrote: On Fri, Oct 25, 2019 at 4:44 AM Ilya Maximets wrote: There is a race window between getting the time and getting the meter lock. This could lead to situation where the thread with larger current time (this thread called time_{um}sec() later than others)

Re: [ovs-dev] [PATCH] dpif-netdev: Do not mix recirculation depth into RSS hash itself.

On 24.10.2019 15:36, Jan Scheurich wrote: Even simpler solution to the problem. Acked-by: Jan Scheurich BR, Jan -Original Message- From: Ilya Maximets Sent: Thursday, 24 October, 2019 14:32 To: ovs-dev@openvswitch.org Cc: Ian Stokes ; Kevin Traynor ; Jan Scheurich ;

Re: [ovs-dev] [PATCH 3/3] Limit the number of generated conjunctions.

On Mon, Oct 28, 2019 at 5:29 PM Dumitru Ceara wrote: > > On Mon, Oct 28, 2019 at 5:07 PM Mark Michelson wrote: > > > > On 10/28/19 10:37 AM, Dumitru Ceara wrote: > > > On Mon, Oct 28, 2019 at 1:46 PM Mark Michelson > > > wrote: > > >> > > >> On 10/28/19 7:46 AM, Dumitru Ceara wrote: > > >>> On

Re: [ovs-dev] [PATCH 3/3] Limit the number of generated conjunctions.

On Mon, Oct 28, 2019 at 5:07 PM Mark Michelson wrote: > > On 10/28/19 10:37 AM, Dumitru Ceara wrote: > > On Mon, Oct 28, 2019 at 1:46 PM Mark Michelson wrote: > >> > >> On 10/28/19 7:46 AM, Dumitru Ceara wrote: > >>> On Fri, Oct 25, 2019 at 11:07 PM Mark Michelson > >>> wrote: > >

Re: [ovs-dev] [PATCH 2/3] Combine conjunctions with identical matches into one flow.

On Mon, Oct 28, 2019, 9:29 PM Mark Michelson wrote: > On 10/28/19 11:33 AM, Numan Siddique wrote: > > On Sat, Oct 26, 2019 at 2:37 AM Mark Michelson > wrote: > >> > >> As stated in previous commits, conjunctive matches have an issue where > >> it is possible to install multiple flows that have

Re: [ovs-dev] [PATCH 3/3] Limit the number of generated conjunctions.

On 10/28/19 10:37 AM, Dumitru Ceara wrote: On Mon, Oct 28, 2019 at 1:46 PM Mark Michelson wrote: On 10/28/19 7:46 AM, Dumitru Ceara wrote: On Fri, Oct 25, 2019 at 11:07 PM Mark Michelson wrote: There is a maximum number of resubmits that can occur during packet processing. Deliberately

Re: [ovs-dev] [PATCH 2/3] Combine conjunctions with identical matches into one flow.

On 10/28/19 11:33 AM, Numan Siddique wrote: On Sat, Oct 26, 2019 at 2:37 AM Mark Michelson wrote: As stated in previous commits, conjunctive matches have an issue where it is possible to install multiple flows that have identical matches. This results in ambiguity, and can lead to features

Re: [ovs-dev] [PATCH 2/3] Combine conjunctions with identical matches into one flow.

On Sat, Oct 26, 2019 at 2:37 AM Mark Michelson wrote: > > As stated in previous commits, conjunctive matches have an issue where > it is possible to install multiple flows that have identical matches. > This results in ambiguity, and can lead to features (such as ACLs) not > functioning properly.

[ovs-dev] [PATCH v4 ovn 2/2] Add DNSSL support to OVN

Introduce the possibility to specify a DNSSL option to Router Advertisement packets. DNS Search list can be specified using 'dnssl' tag in the ipv6_ra_configs column of logical router port table Signed-off-by: Lorenzo Bianconi --- controller/pinctrl.c | 51

[ovs-dev] [PATCH v4 ovn 1/2] Add RDNSS support to OVN

Introduce the possibility to specify a RDNSS option to Router Advertisement packets. DNS IPv6 address can be specified using 'rdnss' tag in the ipv6_ra_configs column of logical router port table Acked-by: Mark Michelson Signed-off-by: Lorenzo Bianconi --- controller/pinctrl.c | 39

Re: [ovs-dev] [PATCH 3/3] Limit the number of generated conjunctions.

On Mon, Oct 28, 2019 at 1:46 PM Mark Michelson wrote: > > On 10/28/19 7:46 AM, Dumitru Ceara wrote: > > On Fri, Oct 25, 2019 at 11:07 PM Mark Michelson wrote: > >> > >> There is a maximum number of resubmits that can occur during packet > >> processing. Deliberately creating a flow table that

[ovs-dev] [PATCH v4 ovn 0/2] add RDNSS/DNSSL support to OVN

Changes since v3: - rely on dp_packet_put to fill rdnss/dnssl addresses in RA packets and remove unused fields Changes since v2: - move option definition in ovn-l7.h - use ovs_16aligned_be32 for option lifetime field Changes since v1: - fix sparse warnings - rebase DNSSL patch on top of RDNSS

[ovs-dev] [PATCH ovn] ovndb-servers.ocf: Change from 'openvswitch' to 'ovn' in default vars.

From: Numan Siddique CC: Aliasgar Ginwala Signed-off-by: Numan Siddique --- utilities/ovndb-servers.ocf | 14 +++--- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/utilities/ovndb-servers.ocf b/utilities/ovndb-servers.ocf index cd4742668..5fe7849ab 100755 ---

Re: [ovs-dev] [PATCH ovn 00/19] OVN Interconnection

On Mon, Oct 21, 2019 at 6:23 AM Han Zhou wrote: > > The series supports interconnecting multiple OVN deployments (e.g. located at > multiple data centers) through logical routers connected with tansit logical > switches with overlay tunnels, managed through OVN control plane. See the >

Re: [ovs-dev] Travis build failures due to base image change to Xenial.

On 09.08.2019 12:23, Ilya Maximets wrote: On 01.08.2019 20:48, Aaron Conole wrote: Ilya Maximets writes: Hi, everyone. I'm trying to fix TravisCI build failures on older branches. Recently they started to change default images from Trusty to Xenial. Not all the repositories affected so far,

Re: [ovs-dev] [PATCH ovs v3 0/2] Introduce dpdkvdpa netdev

On 17.10.2019 13:16, Noa Ezra wrote: There are two approaches to communicate with a guest, using virtIO or SR-IOV. SR-IOV allows working with port representor which is attached to the OVS and a matching VF is given with pass-through to the VM. HW rules can process packets from up-link and direct

Re: [ovs-dev] [PATCH v2 2/2 ovn] OVN: Use ipv4.src and ipv4.dst actions for NAT rules

On Sat, Oct 5, 2019 at 1:45 AM Ankur Sharma wrote: > > For dnat_and_snat rules which are meant to be stateless > instead of using ct_snat/dnat OVN actions, we will use > ipv4.src/ipv4.dst. > > This actions will do 1:1 mapping to inner ip to external ip, > while recalculating the checksums. > >

Re: [ovs-dev] [PATCH v2 1/2 ovn] OVN: ADD nbctl cli to mark a dnat_and_snat rule as stateless

On Thu, Oct 17, 2019 at 10:50 PM Ankur Sharma wrote: > Hi, > > Gentle reminder for reviewing this series. > > Regards, > Ankur > > From: Ankur Sharma > Sent: Friday, October 4, 2019 1:13 PM > To: ovs-dev@openvswitch.org > Cc: Ankur Sharma > Subject: [PATCH v2

Re: [ovs-dev] [PATCH V2] Add offload packets statistics

Thanks for your reply. I already modified some of them, and I think the others need to discuss. V2===>V3: 1. Make the n_offload_packets to be a subset of n_packets, and n_offload_bytes to be a subset of n_bytes. 2. Add a new structure 'dpif_flow_detailed_stats' to store the offload statistics,

Re: [ovs-dev] [dpdk-latest PATCH] netdev-dpdk: add custom vhost statistics to count IRQs

On 28.10.2019 11:14, Eelco Chaudron wrote: On 25 Oct 2019, at 17:06, Ilya Maximets wrote: On 25.10.2019 15:51, Eelco Chaudron wrote:  +static +void vhost_guest_notified(int vid) +{ +    struct netdev_dpdk *dev; + +    ovs_mutex_lock(_mutex); +    LIST_FOR_EACH (dev, list_node, _list) {

Re: [ovs-dev] [PATCH 3/3] Limit the number of generated conjunctions.

On 10/28/19 7:46 AM, Dumitru Ceara wrote: On Fri, Oct 25, 2019 at 11:07 PM Mark Michelson wrote: There is a maximum number of resubmits that can occur during packet processing. Deliberately creating a flow table that might cross this threshold is irresponsible. This commit causes the ofctrl

[ovs-dev] [PATCH V3] Add offload packets statistics

Add argument '-m' or '--more' for command ovs-appctl bridge/dump-flows to display the offloaded packets statistics. The commands display as below: orignal command: ovs-appctl bridge/dump-flows br0 duration=574s, n_packets=1152, n_bytes=110768, priority=0,actions=NORMAL table_id=254,

Re: [ovs-dev] [PATCH 3/3] Limit the number of generated conjunctions.

On Fri, Oct 25, 2019 at 11:07 PM Mark Michelson wrote: > > There is a maximum number of resubmits that can occur during packet > processing. Deliberately creating a flow table that might cross this > threshold is irresponsible. > > This commit causes the ofctrl code to track the maximum width and

Re: [ovs-dev] [PATCHv6] netdev-afxdp: Add need_wakeup supprt.

On 23 Oct 2019, at 23:06, William Tu wrote: The patch adds support for using need_wakeup flag in AF_XDP rings. A new option, use_need_wakeup, is added. When this option is used, it means that OVS has to explicitly wake up the kernel RX, using poll() syscall and wake up TX, using sendto()

Re: [ovs-dev] [PATCHv5] netdev-linux: Detect numa node id.

On 23 Oct 2019, at 23:08, William Tu wrote: > The patch detects the numa node id from the name of the netdev, > by reading the '/sys/class/net//device/numa_node'. > If not available, ex: virtual device, or any error happens, > return numa id 0. Currently only the afxdp netdev type uses it, >

Re: [ovs-dev] [dpdk-latest PATCH] netdev-dpdk: add custom vhost statistics to count IRQs

On 25 Oct 2019, at 17:06, Ilya Maximets wrote: On 25.10.2019 15:51, Eelco Chaudron wrote:  +static +void vhost_guest_notified(int vid) +{ +    struct netdev_dpdk *dev; + +    ovs_mutex_lock(_mutex); +    LIST_FOR_EACH (dev, list_node, _list) { +    if (netdev_dpdk_get_vid(dev) ==

[ovs-dev] [dpdk-latest PATCH] sparse: Remove dpdk network headers copies.

Now that dpdk headers annotates the network header structures with endianness [1], we can get rid of these copies. 1: https://git.dpdk.org/dpdk/commit/?id=7eca7f7fd09d Signed-off-by: David Marchand --- include/sparse/automake.mk | 6 - include/sparse/rte_esp.h | 65 --

[ovs-dev] [PATCH] rhel: openvswitch-fedora.spec.in: Fix output redirect to null device

Add missing slash. Fixes: 0447019df7c6 ("fedora-spec: added systemd post/postun/pre/preun sections") Signed-off-by: Roi Dayan --- rhel/openvswitch-fedora.spec.in | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/rhel/openvswitch-fedora.spec.in

Re: [ovs-dev] [PATCH] tc: Limit the max action number to 16

On 2019-10-28 10:15 AM, Simon Horman wrote: > On Fri, Oct 25, 2019 at 12:19:22PM +0200, Simon Horman wrote: >> On Thu, Oct 24, 2019 at 10:43:00AM +0200, Simon Horman wrote: >>> On Mon, Oct 21, 2019 at 07:01:38AM +, Roi Dayan wrote: On 2019-10-18 1:00 PM, Simon Horman wrote:

Re: [ovs-dev] [PATCH] tc: Limit the max action number to 16

On Fri, Oct 25, 2019 at 12:19:22PM +0200, Simon Horman wrote: > On Thu, Oct 24, 2019 at 10:43:00AM +0200, Simon Horman wrote: > > On Mon, Oct 21, 2019 at 07:01:38AM +, Roi Dayan wrote: > > > > > > > > > On 2019-10-18 1:00 PM, Simon Horman wrote: > > > > On Wed, Oct 16, 2019 at 11:53:52AM

Re: [ovs-dev] [PATCH net-next v4 08/10] net: openvswitch: fix possible memleak on destroy flow-table

On Thu, Oct 24, 2019 at 3:14 PM Pravin Shelar wrote: > > On Tue, Oct 22, 2019 at 7:35 PM Tonghao Zhang > wrote: > > > > On Tue, Oct 22, 2019 at 2:58 PM Pravin Shelar wrote: > > > > ... > > > > > > > > > Sure, I can review it, Can you send the patch inlined in mail? > > > > > > Thanks. > > diff