Re: [ovs-dev] [PATCH v2] netdev-linux: do not touch LAG slaves if master is not attached to ovs.

2022-06-21 Thread Tao Liu
On Tue, Jun 21, 2022 at 06:49:42AM -0700, Marcelo Ricardo Leitner wrote: > Hi, > > On Tue, Jun 21, 2022 at 02:10:39PM +0800, Tao Liu wrote: > > Bond master netdev may be created without a classification type, due > > to routing or tunneling code. > > Can you please elaborate on why is this an

Re: [ovs-dev] [PATCH v1 1/1] datapath-windows: Update layers for multiple tunnels processing

2022-06-21 Thread Alin-Gabriel Serdean
Applied on master! Thank you! Alin. -Original Message- From: dev On Behalf Of Wilson Peng via dev Sent: Thursday, June 2, 2022 9:44 AM To: d...@openvswitch.org Subject: [ovs-dev] [PATCH v1 1/1] datapath-windows: Update layers for multiple tunnels processing From: Wilson Peng While

Re: [ovs-dev] [PATCH v2] datapath-windows: Fix GRE/VxLAN/STT Tunnel RX.

2022-06-21 Thread Alin-Gabriel Serdean
Applied on master! Thank you! Alin -Original Message- From: William Tu Sent: Monday, May 30, 2022 5:55 AM To: d...@openvswitch.org Cc: pweis...@vmware.com; Alin-Gabriel Serdean Subject: [PATCH v2] datapath-windows: Fix GRE/VxLAN/STT Tunnel RX. GRE/Vxlan/STT tunnel RX is broken due to

[ovs-dev] [PATCH net] net: openvswitch: fix parsing of nw_proto for IPv6 fragments

2022-06-21 Thread Rosemarie O'Riorden
When a packet enters the OVS datapath and does not match any existing flows installed in the kernel flow cache, the packet will be sent to userspace to be parsed, and a new flow will be created. The kernel and OVS rely on each other to parse packet fields in the same way so that packets will be

[ovs-dev] [PATCH v4] bond: Improve bond and lacp visibility

2022-06-21 Thread Mike Pattrick
Add additional logging for debug and info level with a focus on code related to bond members coming up, go down, and changing. Several existing log messages were modified to handle sub 1kB log messages with more grace. Instead of reporting 0kB of traffic load shifting from one member to another,

Re: [ovs-dev] [PATCH v5 1/2] handlers: Create additional handler threads when using CPU isolation

2022-06-21 Thread Michael Santana
On Fri, Jun 17, 2022 at 12:10 PM Mike Pattrick wrote: > > On Mon, Jun 6, 2022 at 3:00 PM Michael Santana wrote: > > > > Additional threads are required to service upcalls when we have CPU > > isolation (in per-cpu dispatch mode). The reason additional threads > > are required is because it

[ovs-dev] [PATCH] dpif-netdev-extract-avx512: Protect GCC builtin usage.

2022-06-21 Thread Cian Ferriter
__builtin_constant_p is only available in GCC and only versions >= 4. Use the same "#if __GNUC__ >= 4" check used in other parts of OVS for this builtin. Signed-off-by: Cian Ferriter --- lib/dpif-netdev-extract-avx512.c | 4 1 file changed, 4 insertions(+) diff --git

Re: [ovs-dev] [PATCH v2] tests: Add ovs-dpdk rate limiting unit tests.

2022-06-21 Thread Phelan, Michael
> -Original Message- > From: Phelan, Michael > Sent: Thursday 9 June 2022 11:49 > To: Ilya Maximets ; d...@openvswitch.org; Stokes, Ian > > Cc: maxime.coque...@redhat.com; david.march...@redhat.com; Ryan, > Seamus ; Aaron Conole > Subject: RE: [ovs-dev] [PATCH v2] tests: Add ovs-dpdk

[ovs-dev] [PATCH v3] tests: Add ovs-dpdk rate limiting unit tests.

2022-06-21 Thread Michael Phelan
From: Seamus Ryan This adds 4 new unit tests to the 'check-dpdk' subsystem that will test rate limiting functionality. Signed-off-by: Seamus Ryan Signed-off-by: Michael Phelan Co-authored-by: Michael Phelan --- v3: - Removed NEWS entry. - Added check to catch error if policer fails to

Re: [ovs-dev] [PATCH v2] netdev-linux: do not touch LAG slaves if master is not attached to ovs.

2022-06-21 Thread Marcelo Ricardo Leitner
Hi, On Tue, Jun 21, 2022 at 02:10:39PM +0800, Tao Liu wrote: > Bond master netdev may be created without a classification type, due > to routing or tunneling code. Can you please elaborate on why is this an issue? > > If bond master is not attached to ovs, the ingress block on slaves shoud >

Re: [ovs-dev] [PATCH v7 00/11] Actions Infrastructure + Optimizations

2022-06-21 Thread Eelco Chaudron
Hi Emma, I started reviewing your patch and found some issues that need investigation. With the autovalidator, some tests are failing. But before you can run them, you need to fix your patch 7, as the autovalidator enablement is not working: [wsfd-advnetlab44:~/...DK_v21.11/ovs_github]$ git

Re: [ovs-dev] User space connection tracking benchmarks

2022-06-21 Thread Ilya Maximets
On 6/20/22 23:57, Paolo Valerio wrote: > Ilya Maximets writes: > >> On 6/7/22 11:39, Robin Jarry wrote: >>> Paolo Valerio, Jun 05, 2022 at 19:37: Just a note that may be useful. After some tests, I noticed that establishing e.g. two TCP connections, and leaving the first one idle

[ovs-dev] [PATCH] drop-stats.at: Fix frequent failures of the recursion too deep test.

2022-06-21 Thread Ilya Maximets
The test doesn't wait for old flows being revalidated before sending the second packet. The packet hits old flows and doesn't increase the new drop counter as a result. Solution is to wait for revalidators to clean up old flows. This fixes frequent test failures on CirrusCI. Fixes:

Re: [ovs-dev] [PATCH v4 3/3] system-dpdk: Add unit test for user configured mempools.

2022-06-21 Thread David Marchand
On Wed, Jun 1, 2022 at 6:38 PM Kevin Traynor wrote: > > Test that user configured mempool params have been stored. > > Signed-off-by: Kevin Traynor Reviewed-by: David Marchand > --- > tests/system-dpdk.at | 34 ++ > 1 file changed, 34 insertions(+) > > diff

Re: [ovs-dev] [PATCH v4 2/3] system-dpdk: Split ovsdb creation and vswitchd start.

2022-06-21 Thread David Marchand
On Wed, Jun 1, 2022 at 6:38 PM Kevin Traynor wrote: > > Splitting them allows them to be reused separately. This > is useful for setting some things in ovsdb before vswitchd is > started or DPDK is initialized. > > Signed-off-by: Kevin Traynor Reviewed-by: David Marchand -- David Marchand

Re: [ovs-dev] [PATCH ovn v1 1/3] actions: add sample action

2022-06-21 Thread Adrian Moreno
On 6/16/22 22:31, Mark Michelson wrote: Hi Adrian, just a small comment below. On 6/13/22 12:10, Adrian Moreno wrote: sample ovn action encodes into the OFPACT_SAMPLE ovs action. OVN action allows the following parameters: - obs_domain_id: 8-bit integer that identifies the sampling

Re: [ovs-dev] [PATCH ovn v1 3/3] northd: add drop sampling

2022-06-21 Thread Adrian Moreno
On 6/16/22 22:32, Mark Michelson wrote: Hi Adrian, I have some nitpicky comments below. Thanks Mark, I'll addresss them in the next version of this series. On 6/13/22 12:10, Adrian Moreno wrote: Two new options are added to NB_Global table that enable drop sampling by specifying the

[ovs-dev] [PATCH] tests: Add OVS-DPDK MTU unit tests.

2022-06-21 Thread Michael Phelan
This adds 8 new unit tests to the 'check-dpdk' subsystem that will test Maximum Transmission Unit (MTU) functionality. Signed-off-by: Michael Phelan --- tests/system-dpdk.at | 344 +++ 1 file changed, 344 insertions(+) diff --git a/tests/system-dpdk.at

[ovs-dev] [PATCH] tests: Add OVS-DPDK QoS unit tests

2022-06-21 Thread Michael Phelan
This adds 4 new unit tests to the 'check-dpdk' subsystem that will test Quality of Service (QoS) functionality. Signed-off-by: Michael Phelan --- tests/system-dpdk.at | 141 +++ 1 file changed, 141 insertions(+) diff --git a/tests/system-dpdk.at

Re: [ovs-dev] [PATCH v4 1/3] netdev-dpdk: Add shared mempool config.

2022-06-21 Thread David Marchand
On Wed, Jun 1, 2022 at 6:38 PM Kevin Traynor wrote: > > Mempools may currently be shared between DPDK ports based > on port MTU and NUMA. With some hint from the user we can > increase the sharing on MTU and hence reduce memory > consumption in many cases. > > For example, a port with MTU 9000,

Re: [ovs-dev] [v5 8/8] dpif-netlink: Offloading meter to tc police action

2022-06-21 Thread Jianbo Liu via dev
On Mon, 2022-06-20 at 12:18 +0200, Eelco Chaudron wrote: > On 27 May 2022, at 11:00, Jianbo Liu wrote: > > > OVS meters are created in advance and openflow rules refer to them > > by > > their unique ID. New tc_police API is used to offload them. By > > calling > > the API, police actions are

[ovs-dev] [PATCH v2] netdev-linux: do not touch LAG slaves if master is not attached to ovs.

2022-06-21 Thread Tao Liu
Bond master netdev may be created without a classification type, due to routing or tunneling code. If bond master is not attached to ovs, the ingress block on slaves shoud not be updated. Simple reproducer: ip a add 10.1.1.1/30 dev bond0 ip l set net3 master bond0 tc q ls dev net3 Fixes: