Hello:
This patch was applied to netdev/net.git (main)
by Jakub Kicinski :
On Mon, 22 Apr 2024 05:37:17 -0400 you wrote:
> Since kfree_rcu, which is called in the hlist_for_each_entry_rcu traversal
> of ovs_ct_limit_exit, is not part of the RCU read critical section, it
> is possible that the
On Wed, 24 Apr 2024 17:44:05 +0100 Simon Horman wrote:
> I have recently been exercising the Open vSwitch kernel selftests,
> using vng,
Speaking of ovs tests, we currently don't run them in CI (and suffer
related skips in pmtu.sh) because Amazon Linux doesn't have ovs
packaged and building it
Hello:
This patch was applied to netdev/net-next.git (main)
by Jakub Kicinski :
On Tue, 23 Apr 2024 15:37:51 +0800 you wrote:
> dev_get_by_name will provide a reference on the netdev. So ensure that
> the reference of netdev is released after completed.
>
> Fixes: 2540088b836f ("net:
On Wed, Apr 24, 2024 at 5:14 PM Mark Michelson wrote:
>
> Hi Numan,
>
> I haven't done a full review of this yet, but I figured I'd give some
> initial feedback from what I had looked at.
>
> At a high level, this is missing documentation in ovn-nb.xml for the new
> "overlay_provider_network"
On Wed, Apr 24, 2024 at 5:57 AM Naveen Yerramneni <
naveen.yerramn...@nutanix.com> wrote:
> NB SCHEMA CHANGES
> -
> 1. New DHCP_Relay table
> "DHCP_Relay": {
> "columns": {
> "name": {"type": "string"},
> "servers": {"type":
On Wed, Apr 24, 2024 at 5:57 AM Naveen Yerramneni <
naveen.yerramn...@nutanix.com> wrote:
> Added changes in pinctrl to process DHCP Relay opcodes:
> - ACTION_OPCODE_DHCP_RELAY_REQ_CHK: For request packets
> - ACTION_OPCODE_DHCP_RELAY_RESP_CHK: For response packet
>
> Signed-off-by: Naveen
On Wed, Apr 24, 2024 at 5:56 AM Naveen Yerramneni <
naveen.yerramn...@nutanix.com> wrote:
> NEW OVN ACTIONS
> ---
> 1. dhcp_relay_req_chk(, )
>- This action executes on the source node on which the DHCP request
> originated.
>- This action relays the DHCP request
On Wed, Apr 24, 2024 at 9:21 AM Naveen Yerramneni <
naveen.yerramn...@nutanix.com> wrote:
>
>
> > On 05-Apr-2024, at 9:08 PM, Numan Siddique wrote:
> >
> > CAUTION: External Email
> >
> >
> > On Wed, Mar 20, 2024 at 10:40 AM Naveen Yerramneni <
> naveen.yerramn...@nutanix.com> wrote:
> > >
> > >
On 2024-04-24 18:37 +0100, Simon Horman wrote:
> On Wed, Apr 24, 2024 at 05:44:05PM +0100, Simon Horman wrote:
> > Hi Aaron, Jakub, all,
> >
> > I have recently been exercising the Open vSwitch kernel selftests,
> > using vng, something like this:
> >
> >
Hi Numan,
I haven't done a full review of this yet, but I figured I'd give some
initial feedback from what I had looked at.
At a high level, this is missing documentation in ovn-nb.xml for the new
"overlay_provider_network" option. There should also be a NEWS entry for
the new
Signed-off-by: Adrian Moreno
---
tests/system-common-macros.at| 4 +++
tests/system-offloads-traffic.at | 53
2 files changed, 57 insertions(+)
diff --git a/tests/system-common-macros.at b/tests/system-common-macros.at
index 2a68cd664..860d6a8c9 100644
---
Offload the sample action if it contains psample information by creating
a tc "sample" action with the user cookie inside the action's cookie.
Avoid using the "sample" action's cookie to store the ufid.
Signed-off-by: Adrian Moreno
---
include/linux/automake.mk| 5 +-
Signed-off-by: Adrian Moreno
---
ofproto/ofproto-dpif-psample.c | 59 ++
ofproto/ofproto-dpif-psample.h | 1 +
ofproto/ofproto-dpif.c | 1 +
3 files changed, 61 insertions(+)
diff --git a/ofproto/ofproto-dpif-psample.c b/ofproto/ofproto-dpif-psample.c
This simple program reads from psample and prints the packets to stdout.
It's useful for quickly collecting sampled packets.
Signed-off-by: Adrian Moreno
---
Documentation/automake.mk | 1 +
Documentation/conf.py | 2 +
Documentation/ref/index.rst | 1 +
Add a cache entry type for psample objects.
Store both the dpif_psample reference and the collector_set_id so we can
quickly find the particular exporter.
Using that mechanism, account for packet and byte statistics.
Signed-off-by: Adrian Moreno
---
ofproto/ofproto-dpif-psample.c | 20
Add a new resource in ofproto-dpif and the corresponding API in
ofproto_provider.h to represent and change psample configuration.
Signed-off-by: Adrian Moreno
---
ofproto/automake.mk| 2 +
ofproto/ofproto-dpif-psample.c | 167 +
When a OFP_SAMPLE action is xlated and a dpif_psample object has been
configured (via Flow_Sample_Collector_Set table) with the same
collector_set_id, add psample information to the odp sample action.
Signed-off-by: Adrian Moreno
---
ofproto/ofproto-dpif-psample.c | 20 +
In preparation to allowing certain actions to have a cookie that does
not represent the entire flow, rename flower->act_cookie to
flower->flow_cookie.
This patch does not introduce any behavioral change, it's just a
variable renaming.
Signed-off-by: Adrian Moreno
---
lib/netdev-offload-tc.c |
The new odp sample attributes allow userspace to specify a group_id and
user-defined cookie to be passed down to psample.
Add support for parsing and formatting such action.
Signed-off-by: Adrian Moreno
---
include/linux/openvswitch.h | 49 +---
lib/odp-execute.c| 3 +
Add a psample_group field to the Flow Sample Collector Set table and use
it to configure the psample ofproto layer.
Signed-off-by: Adrian Moreno
---
vswitchd/bridge.c | 54 ++
vswitchd/vswitch.ovsschema | 7 -
vswitchd/vswitch.xml | 32
This is the userspace counterpart of the work being done in the kernel
[1]. Sending it as RFC to get some early feedback on the overall
solution.
** Problem description **
Currently, OVS supports several observability features, such as
per-bridge IPFIX, per-flow IPFIX and sFlow. However, given
Only kernel datapath supports psample so check that the datapath is not
userspace and that it accepts the new attributes.
Signed-off-by: Adrian Moreno
---
ofproto/ofproto-dpif.c | 59 ++
ofproto/ofproto-dpif.h | 6 -
2 files changed, 64
Benjamin Poirier writes:
> On 2024-04-24 18:37 +0100, Simon Horman wrote:
>> On Wed, Apr 24, 2024 at 05:44:05PM +0100, Simon Horman wrote:
>> > Hi Aaron, Jakub, all,
>> >
>> > I have recently been exercising the Open vSwitch kernel selftests,
>> > using vng, something like this:
>> >
>> >
Simon Horman writes:
> Hi Aaron, Jakub, all,
>
> I have recently been exercising the Open vSwitch kernel selftests,
> using vng, something like this:
>
> TESTDIR="tools/testing/selftests/net/openvswitch"
>
> vng -v --run . --user root --cpus 2 \
> --overlay-rwdir
On Wed, Apr 24, 2024 at 05:44:05PM +0100, Simon Horman wrote:
> Hi Aaron, Jakub, all,
>
> I have recently been exercising the Open vSwitch kernel selftests,
> using vng, something like this:
>
> TESTDIR="tools/testing/selftests/net/openvswitch"
>
> vng -v --run . --user root
Hi Aaron, Jakub, all,
I have recently been exercising the Open vSwitch kernel selftests,
using vng, something like this:
TESTDIR="tools/testing/selftests/net/openvswitch"
vng -v --run . --user root --cpus 2 \
--overlay-rwdir "$PWD" -- \
"modprobe
Update:
Seems in upstream 5.4 linux, it only clears vlan_present vs old 4.15 kernel
https://github.com/torvalds/linux/blob/v5.4/net/core/skbuff.c#L5408
int skb_vlan_pop(struct sk_buff *skb)
{
u16 vlan_tci;
__be16 vlan_proto;
int err;
if (likely(skb_vlan_tag_present(skb))) {
Similar to what is already implemented for routed e/w traffic,
introduce pmtud support for e/w traffic between two logical switch ports
connected to the same logical switch, but running on two different
hypervisors.
Acked-by: Mark Michelson
Reported-at: https://issues.redhat.com/browse/FDP-524
Jun Gu writes:
> dev_get_by_name will provide a reference on the netdev. So ensure that
> the reference of netdev is released after completed.
>
> Fixes: 2540088b836f ("net: openvswitch: Check vport netdev name")
> Signed-off-by: Jun Gu
> ---
Thanks!
Reviewed-by: Aaron Conole
Hyunwoo Kim writes:
> Since kfree_rcu, which is called in the hlist_for_each_entry_rcu traversal
> of ovs_ct_limit_exit, is not part of the RCU read critical section, it
> is possible that the RCU grace period will pass during the traversal and
> the key will be free.
>
> To prevent this, it
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, add
missing parsing support for "userspace" action (via
Add support for psample sampling via two new attributes to the
OVS_ACTION_ATTR_SAMPLE action.
OVS_SAMPLE_ATTR_PSAMPLE_GROUP used to pass an integer psample group_id.
OVS_SAMPLE_ATTR_PSAMPLE_COOKIE used to pass a variable-length binary
cookie that will be forwared to psample.
The maximum length
Add sample action support to ovs-dpctl.py.
Signed-off-by: Adrian Moreno
---
.../selftests/net/openvswitch/ovs-dpctl.py| 96 ++-
1 file changed, 95 insertions(+), 1 deletion(-)
diff --git a/tools/testing/selftests/net/openvswitch/ovs-dpctl.py
** Background **
Currently, OVS supports several packet sampling mechanisms (sFlow,
per-bridge IPFIX, per-flow IPFIX). These end up being translated into a
userspace action that needs to be handled by ovs-vswitchd's handler
threads only to be forwarded to some third party application that
will
On Tue, Apr 23, 2024 at 11:33:15AM +0100, Simon Horman wrote:
> On Wed, Apr 17, 2024 at 10:43:11AM +0300, Roi Dayan wrote:
> >
> >
> > On 16/04/2024 18:48, Simon Horman wrote:
> > > On Tue, Apr 16, 2024 at 04:21:48PM +0300, Roi Dayan via dev wrote:
> > >> VLOG_WARN_BUF() is allocating memory for
> On 05-Apr-2024, at 9:08 PM, Numan Siddique wrote:
>
> CAUTION: External Email
>
>
> On Wed, Mar 20, 2024 at 10:40 AM Naveen Yerramneni
> wrote:
> >
> > This patch contains changes to enable DHCP Relay Agent support for
> > overlay subnets.
> >
> > USE CASE:
> > --
>
Currently conntrack uses a single large cmap for all connections stored.
This cmap contains all connections for all conntrack zones which are
completely separate from each other. By separating each zone to its own
cmap we can significantly optimize the performance when using multiple
zones.
The
The current test-conntrack benchmark command runs with multiple threads
against a single conntrack zone. We now add a new benchmark-zones
command that allows us to check the performance between multiple zones.
We in there test the following scenarios for one zone while other zones
also contain
Bleep bloop. Greetings Felix Huettner, 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 lacks whitespace around operator
#104 FILE: lib/conntrack-private.h:203:
Bleep bloop. Greetings Felix Huettner, 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: The subject summary should start with a capital.
WARNING: The subject summary
Introduce bfd and bfd_consumer nodes to northd I-P engine to track bfd
connections and northd static_route/policy_route changes.
Signed-off-by: Lorenzo Bianconi
---
northd/en-lflow.c| 19 +--
northd/en-northd.c | 92 +
northd/en-northd.h | 8 ++
Currently conntrack uses a single large cmap for all connections stored.
This cmap contains all connections for all conntrack zones which are
completely separate from each other. By separating each zone to its own
cmap we can significantly optimize the performance when using multiple
zones.
The
The current test-conntrack benchmark command runs with multiple threads
against a single conntrack zone. We now add a new benchmark-zones
command that allows us to check the performance between multiple zones.
We in there test the following scenarios for one zone while other zones
also contain
References: <20240424095607.129155-4-naveen.yerramn...@nutanix.com>
Bleep bloop. Greetings Naveen Yerramneni, 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
NB SCHEMA CHANGES
-
1. New DHCP_Relay table
"DHCP_Relay": {
"columns": {
"name": {"type": "string"},
"servers": {"type": {"key": "string",
"min": 0,
Added changes in pinctrl to process DHCP Relay opcodes:
- ACTION_OPCODE_DHCP_RELAY_REQ_CHK: For request packets
- ACTION_OPCODE_DHCP_RELAY_RESP_CHK: For response packet
Signed-off-by: Naveen Yerramneni
---
controller/pinctrl.c | 597 ++-
lib/ovn-l7.h
NEW OVN ACTIONS
---
1. dhcp_relay_req_chk(, )
- This action executes on the source node on which the DHCP request
originated.
- This action relays the DHCP request coming from client to the server.
Relay-ip is used to update GIADDR in the DHCP header.
2.
This patch contains changes to enable DHCP Relay Agent support for overlay
subnets.
USE CASE:
--
- Enable IP address assignment for overlay subnets from the centralized
DHCP server present in the underlay network.
PREREQUISITES
--
- Logical
This option is applicable only if the lsp is of default 'type'
i.e. type=empty_string (which is a VM (or VIF) interface) and the
lsp also has 'unknown' addresses configured.
If lsp option 'force_fdb_lookup' is set to true, mac addresses
of the lsp (if configured) are not installed in the l2 lookup
49 matches
Mail list logo