Re: [ovs-dev] [PATCH] appveyor: Fix caching of OpenSSL installer.

On 6/11/24 00:09, Alin Serdean wrote: > Acked-by: Alin-Gabriel Serdean Thanks! Applied. Best regards, Ilya Maximets. > > On Mon, Jun 10, 2024 at 11:18 PM Ilya Maximets > wrote: > > Apparently, if the cache dependency is specified, the cache folder > is

Re: [ovs-dev] [PATCH net-next v2 9/9] selftests: openvswitch: add emit_sample test

Adrian Moreno writes: > Add a test to verify sampling packets via psample works. > > In order to do that, create a subcommand in ovs-dpctl.py to listen to > on the psample multicast group and print samples. > > In order to also test simultaneous sFlow and psample actions and > packet truncation,

Re: [ovs-dev] [PATCH net-next v2 6/9] net: openvswitch: store sampling probability in cb.

Adrian Moreno writes: > The behavior of actions might not be the exact same if they are being > executed inside a nested sample action. Store the probability of the > parent sample action in the skb's cb area. What does that mean? > Use the probability in emit_sample to pass it down to

Re: [ovs-dev] [PATCH v3] netdev-dpdk: Use LSC interrupt mode.

On 6/14/24 17:08, David Marchand wrote: > Querying link status may get delayed for an undeterministic (long) time > with mlx5 ports. This is a consequence of the mlx5 driver calling ethtool > kernel API and getting stuck on the kernel RTNL lock while some other > operation is in progress under

Re: [ovs-dev] [PATCH net-next v2 7/9] net: openvswitch: do not notify drops inside sample

On Mon, Jun 03, 2024 at 08:56:41PM +0200, Adrian Moreno wrote: > The OVS_ACTION_ATTR_SAMPLE action is, in essence, > observability-oriented. > > Apart from some corner case in which it's used a replacement of clone() > for old kernels, it's really only used for sFlow, IPFIX and now, > local

Re: [ovs-dev] [PATCH net-next v2 3/9] net: psample: skip packet copy if no listeners

On Mon, Jun 03, 2024 at 08:56:37PM +0200, Adrian Moreno wrote: > If nobody is listening on the multicast group, generating the sample, > which involves copying packet data, seems completely unnecessary. > > Return fast in this case. > > Signed-off-by: Adrian Moreno Reviewed-by: Simon Horman

Re: [ovs-dev] [PATCH net-next v2 2/9] net: sched: act_sample: add action cookie to sample

On Mon, Jun 03, 2024 at 08:56:36PM +0200, Adrian Moreno wrote: > If the action has a user_cookie, pass it along to the sample so it can > be easily identified. > > Signed-off-by: Adrian Moreno Reviewed-by: Simon Horman ___ dev mailing list

Re: [ovs-dev] [PATCH net-next v2 1/9] net: psample: add user cookie

On Mon, Jun 03, 2024 at 08:56:35PM +0200, Adrian Moreno wrote: > Add a user cookie to the sample metadata so that sample emitters can > provide more contextual information to samples. > > If present, send the user cookie in a new attribute: > PSAMPLE_ATTR_USER_COOKIE. > > Signed-off-by: Adrian

Re: [ovs-dev] [PATCH net-next v2 5/9] net: openvswitch: add emit_sample action

On Mon, Jun 03, 2024 at 08:56:39PM +0200, Adrian Moreno wrote: > Add support for a new action: emit_sample. > > This action accepts a u32 group id and a variable-length cookie and uses > the psample multicast group to make the packet available for > observability. > > The maximum length of the

Re: [ovs-dev] [PATCH net-next v2 4/9] net: psample: allow using rate as probability

On Mon, Jun 03, 2024 at 08:56:38PM +0200, Adrian Moreno wrote: > Although not explicitly documented in the psample module itself, the > definition of PSAMPLE_ATTR_SAMPLE_RATE seems inherited from act_sample. > > Quoting tc-sample(8): > "RATE of 100 will lead to an average of one sampled packet

Re: [ovs-dev] [PATCH net-next 2/2] selftests: openvswitch: set value to nla flags

Adrián Moreno writes: > On Mon, Jun 03, 2024 at 03:02:46PM GMT, Aaron Conole wrote: >> Adrian Moreno writes: >> >> > Netlink flags, although they don't have payload at the netlink level, >> > are represented as having a "True" value in pyroute2. >> > >> > Without it, trying to add a flow with a

Re: [ovs-dev] [RFC net-next 6/7] selftests: net: Use the provided dpctl rather than the vswitchd for tests.

Hi Stefano, Thanks for the review! Stefano Brivio writes: > On Thu, 13 Jun 2024 14:13:32 -0400 > Aaron Conole wrote: > >> The current pmtu test infrastucture requires an installed copy of the >> ovs-vswitchd userspace. This means that any automated or constrained >> environments may not have

Re: [ovs-dev] [PATCH v2] netdev-dpdk: Use LSC interrupt mode.

On Fri, Jun 14, 2024 at 10:17 AM Robin Jarry wrote: > > David Marchand, Jun 14, 2024 at 08:48: > > Querying link status may get delayed for an undeterministic (long) time > > with mlx5 ports. This is a consequence of the mlx5 driver calling ethtool > > kernel API and getting stuck on the kernel

Re: [ovs-dev] [PATCH v2] netdev-dpdk: Use LSC interrupt mode.

On Fri, Jun 14, 2024 at 10:40 AM Robin Jarry wrote: > > David Marchand, Jun 14, 2024 at 08:48: > > diff --git a/NEWS b/NEWS > > index 5ae0108d55..1e19beb793 100644 > > --- a/NEWS > > +++ b/NEWS > > @@ -9,6 +9,9 @@ Post-v3.3.0 > > https://github.com/openvswitch/ovs.git > > - DPDK: > >

[ovs-dev] [PATCH v3] netdev-dpdk: Use LSC interrupt mode.

Querying link status may get delayed for an undeterministic (long) time with mlx5 ports. This is a consequence of the mlx5 driver calling ethtool kernel API and getting stuck on the kernel RTNL lock while some other operation is in progress under this lock. One impact for long link status query

Re: [ovs-dev] [PATCH v2] netdev-dpdk: Use LSC interrupt mode.

On Fri, Jun 14, 2024 at 2:48 AM David Marchand wrote: > > Querying link status may get delayed for an undeterministic (long) time > with mlx5 ports. This is a consequence of the mlx5 driver calling ethtool > kernel API and getting stuck on the kernel RTNL lock while some other > operation is in

Re: [ovs-dev] [PATCH] dpif-netdev: Disable XPS (Transmit Packet Steering) for non-pmd ports.

On 6/13/24 12:15, Eli Britstein wrote: > > >> -Original Message- >> From: Ilya Maximets >> Sent: Monday, 10 June 2024 15:53 >> To: Roi Dayan ; d...@openvswitch.org >> Cc: Eli Britstein ; Maor Dickman ; >> i.maxim...@ovn.org >> Subject: Re: [ovs-dev] [PATCH] dpif-netdev: Disable XPS

Re: [ovs-dev] [PATCH] ofp-prop: Fix unaligned 128 bit access.

On 6/12/24 20:12, Mike Pattrick wrote: > On Wed, Jun 12, 2024 at 9:50 AM Ales Musil wrote: > >> >> >> On Wed, Jun 12, 2024 at 3:32 PM Mike Pattrick wrote: >> >>> When compiling with '-fsanitize=address,undefined', the "ovs-ofctl >>> ct-flush" test will yield the following undefined behavior

Re: [ovs-dev] [BUG][meter] ovs crash when add meter openflow

On 6/14/24 13:33, Simon Jones wrote: > ``` > Date: Fri Jun 14 19:25:43 2024 +0800 > > bugfix of meter tc crash. > > diff --git a/lib/netdev-offload.c b/lib/netdev-offload.c > index 9fde5f7a9..d08c5a35f 100644 > --- a/lib/netdev-offload.c > +++ b/lib/netdev-offload.c > @@ -186,7 +186,6 @@

[ovs-dev] [PATCH] vswitchd: Only lock pages that are faulted in.

The main purpose of locking the memory is to ensure that OVS can keep doing what it did before in case of increased memory pressure, e.g., during VM ingest / migration. Fulfilling this requirement can be achieved without locking all the allocated memory, but only the pages already accessed in the

Re: [ovs-dev] [BUG][meter] ovs crash when add meter openflow

``` Date: Fri Jun 14 19:25:43 2024 +0800 bugfix of meter tc crash. diff --git a/lib/netdev-offload.c b/lib/netdev-offload.c index 9fde5f7a9..d08c5a35f 100644 --- a/lib/netdev-offload.c +++ b/lib/netdev-offload.c @@ -186,7 +186,6 @@ netdev_assign_flow_api(struct netdev *netdev)

Re: [ovs-dev] [BUG][meter] ovs crash when add meter openflow

Maybe reason is this: ``` @netdev_offload_tc and @netdev_offload_dpdk will always be register. Then the @meter_set api will be called. The @meter_set could be called only after @init_flow_api, but the BUG happens when @meter_set is called before @init_flow_api is called. Check these code: static

Re: [ovs-dev] [PATCH v2] ofproto: Add upcall/dump-ufid-rules command to map UFIDs to OpenFlow.

On 6/13/24 23:01, Mike Pattrick wrote: > On Tue, Jun 4, 2024 at 7:29 AM Dumitru Ceara wrote: >> >> It improves the debugging experience if we can easily get a list of >> OpenFlow rules and groups that contribute to the creation of a datapath >> flow. >> >> The suggested workflow is: >> a. dump

[ovs-dev] [PATCH v3] ofproto: Add upcall/dump-ufid-rules command to map UFIDs to OpenFlow.

It improves the debugging experience if we can easily get a list of OpenFlow rules and groups that contribute to the creation of a datapath flow. The suggested workflow is: a. dump datapath flows (along with UUIDs), this also prints the core IDs (PMD IDs) when applicable. $ ovs-appctl

Re: [ovs-dev] [RFC net-next 6/7] selftests: net: Use the provided dpctl rather than the vswitchd for tests.

On Thu, 13 Jun 2024 14:13:32 -0400 Aaron Conole wrote: > The current pmtu test infrastucture requires an installed copy of the > ovs-vswitchd userspace. This means that any automated or constrained > environments may not have the requisite tools to run the tests. However, > the pmtu tests

Re: [ovs-dev] [PATCH v2] netdev-dpdk: Use LSC interrupt mode.

David Marchand, Jun 14, 2024 at 08:48: diff --git a/NEWS b/NEWS index 5ae0108d55..1e19beb793 100644 --- a/NEWS +++ b/NEWS @@ -9,6 +9,9 @@ Post-v3.3.0 https://github.com/openvswitch/ovs.git - DPDK: * OVS validated with DPDK 23.11.1. + * Link status changes are now handled via

Re: [ovs-dev] [v4] odp-execute: Fix AVX checksum calculation.

> Op 14 jun 2024 om 10:13 heeft Phelan, Michael het > volgende geschreven: > >  >> >> -Original Message- >> From: Eelco Chaudron >> Sent: Thursday, June 13, 2024 12:45 PM >> To: Finn, Emma ; Phelan, Michael >> >> Cc: Ilya Maximets ; ovs-dev@openvswitch.org; Van >> Haaren, Harry >>

Re: [ovs-dev] [PATCH v2] netdev-dpdk: Use LSC interrupt mode.

David Marchand, Jun 14, 2024 at 08:48: Querying link status may get delayed for an undeterministic (long) time with mlx5 ports. This is a consequence of the mlx5 driver calling ethtool kernel API and getting stuck on the kernel RTNL lock while some other operation is in progress under this lock.

Re: [ovs-dev] [v4] odp-execute: Fix AVX checksum calculation.

> -Original Message- > From: Eelco Chaudron > Sent: Thursday, June 13, 2024 12:45 PM > To: Finn, Emma ; Phelan, Michael > > Cc: Ilya Maximets ; ovs-dev@openvswitch.org; Van > Haaren, Harry > Subject: Re: [v4] odp-execute: Fix AVX checksum calculation. > > > > On 12 Jun 2024, at

[ovs-dev] [BUG] Why snat action NOT work while using non-DPDK port in openflow

Hi all, I'm using ovs-dpdk(ovs 2.17.2, dpdk 21.11.1). Now I found a bug: Why snat action NOT work while using non-DPDK port in openflow. 1. how to reproduce ``` ### add bridge and port link this [root@bogon ~]# ovs-vsctl show 272472f6-3ab9-4cd5-b1dd-76a63f0e6127 Bridge brp0

[ovs-dev] [PATCH v2] netdev-dpdk: Use LSC interrupt mode.

Querying link status may get delayed for an undeterministic (long) time with mlx5 ports. This is a consequence of the mlx5 driver calling ethtool kernel API and getting stuck on the kernel RTNL lock while some other operation is in progress under this lock. One impact for long link status query