Re: [ovs-dev] [PATCH v3] dpif-netdev: Allow PMD auto load balance with cross-numa.

Hello Kevin Thanks for the review of the results. It seems the tables sent from gmail are not formatted correctly, as you have suggested will use plain text or mail client that sends in plain text, so that things are formatted the right way. On Wed, May 5, 2021 at 6:02 PM Kevin Traynor wrote: >

Re: [ovs-dev] [PATCH v8 ovn] ovn-northd: introduce new allow-stateless ACL verb

Bleep bloop. Greetings Ihar Hrachyshka, 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. build: /bin/python3 ./build-aux/dpdkstrip.py | \ sed \ -e

[ovs-dev] [PATCH v8 ovn] ovn-northd: introduce new allow-stateless ACL verb

For allow-stateless ACLs, bypass connection tracking by avoiding setting ct hints for matching traffic. Avoid sending all traffic to ct when a stateful ACL is present. === Reusing an existing 'allow' verb for stateless matching would have its drawbacks, specifically, when it comes to backwards

Re: [ovs-dev] [PATCH ovn v6] ovn-controller: Split logical flow and phsyical flow processing.

On Thu, Apr 29, 2021 at 8:10 AM Numan Siddique wrote: > > On Mon, Apr 26, 2021 at 7:18 PM Han Zhou wrote: > > > > On Mon, Apr 12, 2021 at 6:23 AM wrote: > > > > > > From: Numan Siddique > > > > > > Presently, the 'flow_output' engine node recomputes physical > > > flows by calling

Re: [ovs-dev] [PATCH v1 0/8] RCU: Add blocking mode for debugging

On Thu, May 06, 2021 at 12:37:36AM +0200, Gaëtan Rivet wrote: > On Wed, May 5, 2021, at 21:36, Ben Pfaff wrote: > > On Wed, Apr 28, 2021 at 01:03:24AM +0200, Gaetan Rivet wrote: > > > This series adds a compilation option that changes the behavior of the RCU > > > module. Once enabled, RCU

Re: [ovs-dev] [PATCH 0/7] OVSDB 2-Tier deployment.

On Thu, May 06, 2021 at 12:34:00AM +0200, Ilya Maximets wrote: > On 5/5/21 11:40 PM, Ben Pfaff wrote: > > I think that this series has the details that one needs to understand it > > at a high level, but I didn't see a high-level overview of its intended > > use. I think the idea is that, for a

Re: [ovs-dev] [PATCH v1 0/8] RCU: Add blocking mode for debugging

On Wed, May 5, 2021, at 21:36, Ben Pfaff wrote: > On Wed, Apr 28, 2021 at 01:03:24AM +0200, Gaetan Rivet wrote: > > This series adds a compilation option that changes the behavior of the RCU > > module. Once enabled, RCU reclamation by user threads becomes blocking until > > the RCU threads has

Re: [ovs-dev] [PATCH 0/7] OVSDB 2-Tier deployment.

On 5/5/21 11:40 PM, Ben Pfaff wrote: > On Sat, May 01, 2021 at 02:55:41AM +0200, Ilya Maximets wrote: >> Replication can be used to scale out read-only access to the database. >> But there are clients that are not read-only, but read-mostly. >> One of the main examples is ovn-controller that

Re: [ovs-dev] [PATCH 0/7] OVSDB 2-Tier deployment.

On Sat, May 01, 2021 at 02:55:41AM +0200, Ilya Maximets wrote: > Replication can be used to scale out read-only access to the database. > But there are clients that are not read-only, but read-mostly. > One of the main examples is ovn-controller that mostly monitors > updates from the Southbound

Re: [ovs-dev] [PATCH v2] flow: Read recirculation depth once for whole batch in miniflow_extract

Balazs Nemeth writes: > The call to recirc_depth_get involves accessing a TLS value which is > slower than accessing the stack. > > Signed-off-by: Balazs Nemeth > --- LGTM, Acked-by: Paolo Valerio ___ dev mailing list d...@openvswitch.org

Re: [ovs-dev] [PATCH] lib: Introduce netlink-devlink library

Bleep bloop. Greetings Frode Nordahl, 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. checkpatch: ERROR: Inappropriate spacing in pointer declaration WARNING: Line lacks whitespace around

[ovs-dev] [PATCH] lib: Introduce netlink-devlink library

The devlink interface was introduced [0] in the Linux 4.6 time frame and has since gained traction among multiple hardware vendors. The devlink-port [1] and devlink-info[2] interfaces are particularly useful for managing NICs connected to multiple distinct CPUs such as SmartNICs. In such a

Re: [ovs-dev] [PATCH v1 0/8] RCU: Add blocking mode for debugging

On Wed, Apr 28, 2021 at 01:03:24AM +0200, Gaetan Rivet wrote: > This series adds a compilation option that changes the behavior of the RCU > module. Once enabled, RCU reclamation by user threads becomes blocking until > the RCU threads has executed the scheduled callbacks. It's a great idea to

Re: [ovs-dev] [PATCH V6 00/13] Netdev vxlan-decap offload

On Wed, Apr 21, 2021, at 10:20, Kovacevic, Marko wrote: > > On 4/7/2021 12:10 PM, Sriharsha Basavapatna wrote: > > > > > > On Sun, Apr 4, 2021 at 3:25 PM Eli Britstein > > > wrote: > > > > > > VXLAN decap in OVS-DPDK configuration consists of two flows: > > > F1:

Re: [ovs-dev] [PATCH ovn v7 4/5] pinctrl: Add Chassis MAC_Bindings for all router addresses.

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. checkpatch: WARNING: Line is 84 characters long (recommended limit is 79) #368 FILE:

Re: [ovs-dev] [PATCH ovn v7 3/5] northd: Save all router addresses in Port_Bindings

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. checkpatch: WARNING: Line is 83 characters long (recommended limit is 79) #252 FILE:

[ovs-dev] [PATCH ovn v7 5/5] northd: Flood ARPs to routers for "unreachable" addresses.

Previously, ARP TPAs were filtered down only to "reachable" addresses. Reachable addresses are all router interface addresses, as well as NAT external addresses and load balancer VIPs that are within the subnet handled by a router's port. However, it is possible that in some configurations, CMSes

[ovs-dev] [PATCH ovn v7 3/5] northd: Save all router addresses in Port_Bindings

northd currently stores logical switch port's configured "nat-addresses" in the southbound Port_Binding "nat_addresses" column. This allows for explicit configuration of which NAT addresses should be broadcast in gARP messages by OVN. This adds a similar column, "router_addresses" to the

[ovs-dev] [PATCH ovn v7 4/5] pinctrl: Add Chassis MAC_Bindings for all router addresses.

Previous behavior of ovn-controller was to only create MAC bindings for the same addresses for which we would send gARPs. That is: * A logical switch has its "nat_addresses" configured. * That logical switch has a localnet port. This makes sense for gARPs: we only want to send gARPs for

[ovs-dev] [PATCH ovn v7 1/5] northd: Swap src and dst eth addresses in router egress loop.

When a hairpin scenario is detected (i.e. egressing on gateway router port to a NAT external address), we loop back to the ingress router pipeline and swap the inport and outport. This is intended to allow for the traffic to get DNATted. In practice, though, the ethernet destination has not been

[ovs-dev] [PATCH ovn v7 2/5] ovn-sb: Remove redundant "nat-addresses" information from Port_Binding.

Signed-off-by: Mark Michelson --- ovn-sb.xml | 10 -- 1 file changed, 10 deletions(-) diff --git a/ovn-sb.xml b/ovn-sb.xml index 258a12b4e..b853a5031 100644 --- a/ovn-sb.xml +++ b/ovn-sb.xml @@ -3015,16 +3015,6 @@ tcp.flags = RST; The chassis in which the port resides.

[ovs-dev] [PATCH ovn v7 0/5] ARP and Floating IP Fixes

This patch series aims to fix issues seen in OpenStack deployments when floating IPs were assigned to routers, and those floating IPs were not part of any subnet configured on that router. Originally, this was a two patch series but it has bloomed into a 5 patch series. Patch 1 fixes the

Re: [ovs-dev] [PATCH v3 14/28] seq-pool: Module for faster ID generation

On Mon, May 3, 2021, at 14:48, Ilya Maximets wrote: > On 5/1/21 4:49 PM, Gaëtan Rivet wrote: > > On Fri, Apr 30, 2021, at 19:31, Ilya Maximets wrote: > >> On 4/25/21 1:55 PM, Gaetan Rivet wrote: > >>> The current id-pool module is slow to allocate the > >>> next valid ID, and can be optimized when

Re: [ovs-dev] [PATCH v2] tests: Add PMD auto load balance unit tests.

Ping. Any objection to adding these unit tests? On 16/03/2021 15:45, Kevin Traynor wrote: > These tests focus on enabling/disabling and user parameters. > > Co-Authored-by: David Marchand > Signed-off-by: David Marchand > Signed-off-by: Kevin Traynor > > --- > v2: > - Remove above max

Re: [ovs-dev] [PATCH v3] dpif-netdev: Allow PMD auto load balance with cross-numa.

On 04/05/2021 17:12, Yogananth Subramanian wrote: > Hello Everyone Hi Yogananth, > I was able to verify the patch, below are the observed behavior of OVS > 2.15 release and OVS from git with the patch. > OVS 2.15 release : > > PMD auto balance aborted: > >

Re: [ovs-dev] [PATCH v3] Don't mangle shebangs when building DKMS RPM package

W dniu śro, 05.05.2021 o godzinie 13∶16 +0200, użytkownik Ilya Maximets napisał: > Spółka wpisana do rejestru przedsiębiorców w Sądzie Rejonowym dla m. st. Warszawy, XIII Wydział Gospodarczy, pod numerem KRS: 044577, kapitał zakładowy: 576 854 559 PLN, kapitał opłacony w całości, NIP:

Re: [ovs-dev] [PATCH v3] Don't mangle shebangs when building DKMS RPM package

On 5/5/21 12:53 PM, Guzowski Adrian via dev wrote: > While building the package, some .in files are being subject to shebang > substitution, but the process fails, because given scripts have > placeholders in place of shebangs. In order to fix the issue, don't mangle > shebangs in this specific

[ovs-dev] [PATCH v3] Don't mangle shebangs when building DKMS RPM package

While building the package, some .in files are being subject to shebang substitution, but the process fails, because given scripts have placeholders in place of shebangs. In order to fix the issue, don't mangle shebangs in this specific package. Signed-off-by: Guzowski Adrian ---

Re: [ovs-dev] Add ability to override default Release suffix in RPM packages

W dniu śro, 05.05.2021 o godzinie 12∶33 +0200, użytkownik Timothy Redaelli napisał: > On Tue, 4 May 2021 06:24:33 + Spółka wpisana do rejestru przedsiębiorców w Sądzie Rejonowym dla m. st. Warszawy, XIII Wydział Gospodarczy, pod numerem KRS: 044577, kapitał zakładowy: 576 854 559 PLN,

Re: [ovs-dev] [PATCH v2] rpm: Don't mangle shebangs when building DKMS RPM package

W dniu śro, 05.05.2021 o godzinie 12∶20 +0200, użytkownik Timothy Redaelli napisał: > On Wed, 5 May 2021 05:37:41 + Spółka wpisana do rejestru przedsiębiorców w Sądzie Rejonowym dla m. st. Warszawy, XIII Wydział Gospodarczy, pod numerem KRS: 044577, kapitał zakładowy: 576 854 559 PLN,

Re: [ovs-dev] Add ability to override default Release suffix in RPM packages

On Tue, 4 May 2021 06:24:33 + Guzowski Adrian via dev wrote: > In some cases, like building OvS packages in Jenkins, it may be > useful to set a custom version suffix that will correspond with > job's build number, etc. Currently, version number is explicitly > set to 1. This change adds a

Re: [ovs-dev] [PATCH v2] rpm: Don't mangle shebangs when building DKMS RPM package

On Wed, 5 May 2021 05:37:41 + Guzowski Adrian via dev wrote: > While building the package, some .in files are being subject to shebang > substitution, but the process fails, because given scripts have > placeholders in place of shebangs. In order to fix the issue, don't mangle > shebangs in

Re: [ovs-dev] [PATCH ovn] ofctrl: Don't assert while merging tracked flows.

On Thu, Apr 29, 2021 at 9:25 AM Numan Siddique wrote: > > On Wed, Apr 28, 2021 at 3:10 PM Han Zhou wrote: > > > > On Wed, Apr 28, 2021 at 3:34 AM Numan Siddique wrote: > > > > > > On Wed, Apr 28, 2021 at 3:31 AM Han Zhou wrote: > > > > > > > > Thanks Numan for working on this issue! > > > > >