From: Jan Scheurich
Today dpif-netdev considers PMD threads on a non-local NUMA node for automatic
assignment of the rxqs of a port only if there are no local,non-isolated PMDs.
On typical servers with both physical ports on one NUMA node, this often
leaves the PMDs on the other NUMA node
From: Jan Scheurich
Today dpif-netdev considers PMD threads on a non-local NUMA node for automatic
assignment of the rxqs of a port only if there are no local,non-isolated PMDs.
On typical servers with both physical ports on one NUMA node, this often
leaves the PMDs on the other NUMA node
Bleep bloop. Greetings Anurag Agarwal, 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: Author Jan Scheurich needs to sign off.
WARNING: Unexpected sign-offs from
From: Jan Scheurich
Today dpif-netdev considers PMD threads on a non-local NUMA node for automatic
assignment of the rxqs of a port only if there are no local,non-isolated PMDs.
On typical servers with both physical ports on one NUMA node, this often
leaves the PMDs on the other NUMA node
Bleep bloop. Greetings Anurag Agarwal, 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: Author Anurag Agarwal needs to sign off.
WARNING: Unexpected sign-offs from
Today dpif-netdev considers PMD threads on a non-local NUMA node for automatic
assignment of the rxqs of a port only if there are no local,non-isolated PMDs.
On typical servers with both physical ports on one NUMA node, this often
leaves the PMDs on the other NUMA node under-utilized, wasting
On 6/2/22 21:55, Numan Siddique wrote:
> On Wed, Jun 1, 2022 at 1:12 PM Andreas Karis wrote:
>>
>> Provide options to enforce NAT-T UDP encapsulation. Options are
>> encapsulation=true for libreswan and forceencaps=true for strongswan.
>> This may be required in environments where firewalls drop
On 6/2/22 23:13, Jon Kohler wrote:
> Any takers? I’m hoping I’ve got the right mailing list, as I did see
> the thread get generated on the mailing list website?
Hey, Jon.
Yes, it's the right mailing list, though ovs-dev and dev are
the same thing, so you don't need to send to both.
For the
Any takers? I’m hoping I’ve got the right mailing list, as I did see
the thread get generated on the mailing list website?
> On May 31, 2022, at 10:03 AM, Jon Kohler wrote:
>
>
>
>> On May 26, 2022, at 9:11 PM, Jon Kohler wrote:
>>
>> For netdev_linux_update_via_netlink(), hint to the
On Wed, Jun 1, 2022 at 7:00 PM Ihar Hrachyshka wrote:
>
> This is the remaining piece in multi-chassis series. This new version
> has no flow modification done in pinctrl thread. Instead, I-P engine
> is notified of any newly activated ports, which then triggers pflow
> update for lports.
>
>
On Wed, Jun 1, 2022 at 1:12 PM Andreas Karis wrote:
>
> Provide options to enforce NAT-T UDP encapsulation. Options are
> encapsulation=true for libreswan and forceencaps=true for strongswan.
> This may be required in environments where firewalls drop ESP
> traffic but where NAT-T detection fails
On Thu, Jun 2, 2022 at 11:39 AM Dumitru Ceara wrote:
>
> On 6/2/22 19:22, Han Zhou wrote:
> > On Thu, Jun 2, 2022 at 10:09 AM Mark Michelson
wrote:
> >>
> >> On 6/2/22 11:22, Dumitru Ceara wrote:
> >>> Change the way ovn-controller decides whether it should match on
> >>> ct_mark.natted or
On 6/2/22 14:39, Dumitru Ceara wrote:
On 6/2/22 19:22, Han Zhou wrote:
On Thu, Jun 2, 2022 at 10:09 AM Mark Michelson wrote:
On 6/2/22 11:22, Dumitru Ceara wrote:
Change the way ovn-controller decides whether it should match on
ct_mark.natted or ct_label.natted for hairpin load balancer
On Thu, Jun 2, 2022 at 8:22 AM Dumitru Ceara wrote:
>
> Commit a075230e4a0f ("Use ct_mark for masked access to make flows
> HW-offloading friendly.") started using the ct_mark.blocked and
> ct_mark.ecmp_reply_port ct_label. In usual scenarios this new feature
would
> be picked up when the next
On 6/2/22 19:22, Han Zhou wrote:
> On Thu, Jun 2, 2022 at 10:09 AM Mark Michelson wrote:
>>
>> On 6/2/22 11:22, Dumitru Ceara wrote:
>>> Change the way ovn-controller decides whether it should match on
>>> ct_mark.natted or ct_label.natted for hairpin load balancer traffic.
>>> Until now this was
On Thu, Jun 2, 2022 at 10:09 AM Mark Michelson wrote:
>
> On 6/2/22 11:22, Dumitru Ceara wrote:
> > Change the way ovn-controller decides whether it should match on
> > ct_mark.natted or ct_label.natted for hairpin load balancer traffic.
> > Until now this was done solely based on the
Acked-by: Mark Michelson
On 6/2/22 11:22, Dumitru Ceara wrote:
Commit a075230e4a0f ("Use ct_mark for masked access to make flows
HW-offloading friendly.") started using the ct_mark.blocked and
ct_mark.ecmp_reply_port ct_label. In usual scenarios this new feature would
be picked up when the
On 6/2/22 11:22, Dumitru Ceara wrote:
Change the way ovn-controller decides whether it should match on
ct_mark.natted or ct_label.natted for hairpin load balancer traffic.
Until now this was done solely based on the northd-reported internal
version.
However, to cover the case when OVN central
On 6/1/22 22:53, Gregory Rose wrote:
>
>
> On 5/31/2022 12:22 PM, Frode Nordahl wrote:
>> On Tue, May 31, 2022 at 7:05 PM Ilya Maximets wrote:
>>>
>>> On 5/31/22 17:36, Gregory Rose wrote:
On 5/25/2022 6:47 AM, Flavio Leitner wrote:
>
> Hi Greg,
>
>
> On Mon,
Thanks for the quick fix Ihar. I pushed this to main and branch-22.06
On 6/1/22 23:54, Ihar Hrachyshka wrote:
Fixes fa8c591fa2a7("Support LSP:options:requested-chassis as a list")
Signed-off-by: Ihar Hrachyshka
---
controller/binding.c | 12
1 file changed, 8 insertions(+), 4
On 6/2/22 16:28, Kevin Traynor wrote:
> On 02/06/2022 15:00, Ilya Maximets wrote:
>> On 6/2/22 13:01, Kevin Traynor wrote:
>>> On 01/06/2022 14:30, David Marchand wrote:
On Wed, May 25, 2022 at 1:11 PM Kevin Traynor wrote:
>
> Currently mempools for vhost are being assigned before
Currently mempools for vhost are being assigned before the vhost device
is added. In some cases this may be just reusing an existing mempool but
in others it can require creation of a mempool.
For multi-NUMA, the NUMA info of the vhost port is not known until a
device is added to the port, so on
V2:
- Also implement delayed vhost mempool creation/reuse for single-NUMA
- Added David's RvB, as minor code change from v1 and David had
preference for common single/multi NUMA behaviour
V1 only operated for multi-NUMA, as potentially having to create a mempool
and then recreate it on a
On 6/2/22 16:57, Han Zhou wrote:
> On Thu, Jun 2, 2022 at 7:43 AM Dumitru Ceara wrote:
>>
>> On 6/2/22 16:35, Han Zhou wrote:
>>> On Thu, Jun 2, 2022 at 5:07 AM Dumitru Ceara wrote:
On 6/2/22 04:08, Han Zhou wrote:
> On Wed, Jun 1, 2022 at 11:44 AM Mark Michelson
>>> wrote:
>>
Change the way ovn-controller decides whether it should match on
ct_mark.natted or ct_label.natted for hairpin load balancer traffic.
Until now this was done solely based on the northd-reported internal
version.
However, to cover the case when OVN central components are not the last
ones to be
Commit a075230e4a0f ("Use ct_mark for masked access to make flows
HW-offloading friendly.") started using the ct_mark.blocked and
ct_mark.ecmp_reply_port ct_label. In usual scenarios this new feature would
be picked up when the next stable release becomes available (i.e., 22.06.0).
However, the
Dumitru Ceara (2):
northd: ovn-controller: Use ct_mark.natted only when ct_lb_mark is used.
northd: Use ct_mark.blocked and ecmp_reply_port only when all chassis
support it.
controller/chassis.c | 6 ++
include/ovn/features.h | 5 +-
northd/northd.c| 156
On 6/2/22 05:54, Ihar Hrachyshka wrote:
> Fixes fa8c591fa2a7("Support LSP:options:requested-chassis as a list")
>
> Signed-off-by: Ihar Hrachyshka
> ---
Looks good to me, thanks!
Acked-by: Dumitru Ceara
___
dev mailing list
d...@openvswitch.org
On 2 Jun 2022, at 14:01, Ilya Maximets wrote:
> On 6/1/22 12:37, Eelco Chaudron wrote:
>>
>>
>> On 1 Jun 2022, at 12:14, Roi Dayan wrote:
>>
>>> On 2022-05-16 6:38 PM, Eelco Chaudron wrote:
This patch will properly initialize offload, as it requires the
setting to be enabled before
Hi,
Do you mind sending a patch to the ML adding yourself to the AUTHORS.rst
I can apply it later on.
Thank you,
Alin.
-Original Message-
From: dev On Behalf Of jinjun gao
Sent: Thursday, June 2, 2022 5:38 PM
To: d...@openvswitch.org
Subject: [ovs-dev] Please help add me to AUTHORS
> -Original Message-
> From: Eelco Chaudron
> Sent: Thursday 2 June 2022 15:43
> To: Finn, Emma
> Cc: Van Haaren, Harry ; Amber, Kumar
> ; Stokes, Ian ;
> d...@openvswitch.org
> Subject: Re: [v6 11/11] odp-execute: Add ISA implementation of
> set_masked IPv4 action
>
> On 10 May 2022,
On Thu, Jun 2, 2022 at 7:43 AM Dumitru Ceara wrote:
>
> On 6/2/22 16:35, Han Zhou wrote:
> > On Thu, Jun 2, 2022 at 5:07 AM Dumitru Ceara wrote:
> >>
> >> On 6/2/22 04:08, Han Zhou wrote:
> >>> On Wed, Jun 1, 2022 at 11:44 AM Mark Michelson
> > wrote:
>
> On 6/1/22 13:44, Numan
On Thu, Jun 2, 2022 at 10:34 AM Terry Wilson wrote:
>
> When checking if ovsdb-server is running, ensure that the binary
> we are going to run matches the one actually running with the the
> pid that was in our pidfile.
>
> Signed-off-by: Terry Wilson
Acked-by: Ihar Hrachyshka
> ---
>
On Thu, Jun 2, 2022 at 10:34 AM Terry Wilson wrote:
>
> Since pids can be re-used, it is necessary to check that the
> process that is running with a pid matches the one that we expect.
>
> This adds the ability to optionally pass a 'binary' argument to
> pidfile_is_running, and if it is passed
On 10 May 2022, at 16:22, Emma Finn wrote:
> This commit adds support for the AVX512 implementation of the
> ipv4_set_addrs action as well as an AVX512 implementation of
> updating the checksums.
>
> Signed-off-by: Emma Finn
> ---
> lib/odp-execute-avx512.c | 194
On 6/2/22 16:35, Han Zhou wrote:
> On Thu, Jun 2, 2022 at 5:07 AM Dumitru Ceara wrote:
>>
>> On 6/2/22 04:08, Han Zhou wrote:
>>> On Wed, Jun 1, 2022 at 11:44 AM Mark Michelson
> wrote:
On 6/1/22 13:44, Numan Siddique wrote:
> On Wed, Jun 1, 2022 at 12:52 PM Mark Michelson
>>>
On 10 May 2022, at 16:22, Emma Finn wrote:
> This commit includes infrastructure changes for enabling set_masked_X
> actions and also adds support for the AVX512 implementation of the
> eth_set_addrs action.
>
> Signed-off-by: Emma Finn
> ---
> .../linux/compat/include/linux/openvswitch.h | 2
Bleep bloop. Greetings Terry Wilson, 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 124 characters long (recommended limit is 79)
#36 FILE: utilities/ovn-ctl:52:
Hi guys,
I'm a contributor in datapath-windows part with jinj...@vmware.com. Can you
please help add me to the AUTHORS? Thanks!
Name: Jinjun Gao
Mail: gjin...@gmail.com
--
Best Regards,
Jinjun Gao
___
dev mailing list
d...@openvswitch.org
On 10 May 2022, at 16:22, Emma Finn wrote:
> This commit adds the AVX512 implementation of the
> push_vlan action.
>
> Signed-off-by: Emma Finn
> ---
> lib/odp-execute-avx512.c | 56 ---
> lib/odp-execute-private.c | 1 +
> lib/odp-execute.c | 24
On Thu, Jun 2, 2022 at 5:07 AM Dumitru Ceara wrote:
>
> On 6/2/22 04:08, Han Zhou wrote:
> > On Wed, Jun 1, 2022 at 11:44 AM Mark Michelson
wrote:
> >>
> >> On 6/1/22 13:44, Numan Siddique wrote:
> >>> On Wed, Jun 1, 2022 at 12:52 PM Mark Michelson
> > wrote:
>
> Thanks Dumitru. This
On 10 May 2022, at 16:21, Emma Finn wrote:
> This commit adds the AVX512 implementation of the
> pop_vlan action.
>
> Signed-off-by: Emma Finn
> ---
> lib/odp-execute-avx512.c | 73 ++-
> lib/odp-execute-private.c | 2 +-
> lib/odp-execute-private.h | 2 +-
Since pids can be re-used, it is necessary to check that the
process that is running with a pid matches the one that we expect.
This adds the ability to optionally pass a 'binary' argument to
pidfile_is_running, and if it is passed to match the binary against
/proc/$pid/exe.
Signed-off-by: Terry
When checking if ovsdb-server is running, ensure that the binary
we are going to run matches the one actually running with the the
pid that was in our pidfile.
Signed-off-by: Terry Wilson
---
utilities/ovn-ctl | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/utilities/ovn-ctl
On 10 May 2022, at 16:21, Emma Finn wrote:
> This commit adds the AVX512 implementation of the action functionality.
>
> Usage:
> $ ovs-appctl dpif-netdev/action-impl-set avx512
>
> Signed-off-by: Emma Finn
> Acked-by: Harry van Haaren
> ---
> Documentation/topics/dpdk/bridge.rst | 25
On 02/06/2022 15:00, Ilya Maximets wrote:
On 6/2/22 13:01, Kevin Traynor wrote:
On 01/06/2022 14:30, David Marchand wrote:
On Wed, May 25, 2022 at 1:11 PM Kevin Traynor wrote:
Currently mempools for vhost are being assigned before the vhost device
is added. In some cases this may be just
On 10 May 2022, at 16:21, Emma Finn wrote:
> From: Kumar Amber
>
> This commit adds a new command to allow the user to enable the
> actions autovalidator by default at build time thus allowing for
> running unit test by default.
>
> $ ./configure --enable-actions-default-autovalidator
>
>
On 10 May 2022, at 16:21, Emma Finn wrote:
> This commit adds a new command to allow the user to switch
> the active action implementation at runtime. A probe function
> is executed before switching the implementation, to ensure
> the CPU is capable of running the ISA required.
>
> Usage:
> $
On 10 May 2022, at 16:21, Emma Finn wrote:
> This commit introduced the auto-validation function which
> allows users to compare the batch of packets obtained from
> different action implementations against the linear
> action implementation.
>
> The autovalidator function can be triggered at
On Wed, Jun 1, 2022 at 10:18 PM Ihar Hrachyshka wrote:
>
> On Wed, Jun 1, 2022 at 1:26 PM Terry Wilson wrote:
> >
> > Since pids can be re-used, it is necessary to check that the
> > process that is running with a pid matches the one that we expect.
> >
> > This adds the ability to optionally
On 10 May 2022, at 16:21, Emma Finn wrote:
> This commit removes the pop_vlan action from the large switch
> and creates a separate function for batched processing. A function
> pointer is also added to call the new batched function for the pop_vlan
> action.
>
> Signed-off-by: Emma Finn
>
On 6/2/22 13:01, Kevin Traynor wrote:
> On 01/06/2022 14:30, David Marchand wrote:
>> On Wed, May 25, 2022 at 1:11 PM Kevin Traynor wrote:
>>>
>>> Currently mempools for vhost are being assigned before the vhost device
>>> is added. In some cases this may be just reusing an existing mempool but
On 10 May 2022, at 16:21, Emma Finn wrote:
> This commit introduces the initial infrastructure required to allow
> different implementations for OvS actions. The patch introduces action
> function pointers which allows user to switch between different action
> implementations available. This will
> On 01/06/2022 14:30, David Marchand wrote:
> > On Wed, May 25, 2022 at 1:11 PM Kevin Traynor
> wrote:
> >>
> >> Currently mempools for vhost are being assigned before the vhost
> device
> >> is added. In some cases this may be just reusing an existing mempool
> but
> >> in others it can
On 6/2/22 04:08, Han Zhou wrote:
> On Wed, Jun 1, 2022 at 11:44 AM Mark Michelson wrote:
>>
>> On 6/1/22 13:44, Numan Siddique wrote:
>>> On Wed, Jun 1, 2022 at 12:52 PM Mark Michelson
> wrote:
Thanks Dumitru. This looks good to me.
Acked-by: Mark Michelson
>>>
>>> Thanks
On 6/1/22 12:37, Eelco Chaudron wrote:
>
>
> On 1 Jun 2022, at 12:14, Roi Dayan wrote:
>
>> On 2022-05-16 6:38 PM, Eelco Chaudron wrote:
>>> This patch will properly initialize offload, as it requires the
>>> setting to be enabled before starting ovs-vswitchd (or do a
>>> restart once
On 01/06/2022 14:30, David Marchand wrote:
On Wed, May 25, 2022 at 1:11 PM Kevin Traynor wrote:
Currently mempools for vhost are being assigned before the vhost device
is added. In some cases this may be just reusing an existing mempool but
in others it can require creation of a mempool.
From: Wilson Peng
While testing OVS-windows for multiple IPV6 Geneve tunnels on Windows2019VM,
for the broadcast/mutlicast packets, it needs to flood out via configured
multiple Geneve tunnels. Then in some flow pipeline processing, it may have
at least two tunnel processing in
58 matches
Mail list logo
