Re: [ovs-dev] [PATCH ovn] northd: support HW VTEP with stateful datapath

On Fri, Sep 17, 2021 at 5:56 PM Vladislav Odintsov wrote: > > A packet going from HW VTEP device to VIF port when arrives to > hypervisor chassis should go through LS ingress pipeline to l2_lkp > stage without any match. In l2_lkp stage an output port is > determined and then packet passed to LS

Re: [ovs-dev] [PATCH ovn 1/2] Revert "northd: Don't poll ovsdb before the connection is fully established"

On Fri, Sep 17, 2021 at 5:10 PM Han Zhou wrote: > > > > On Fri, Sep 17, 2021 at 2:04 PM Numan Siddique wrote: > > > > On Fri, Sep 17, 2021 at 5:02 PM Renat Nurgaliyev wrote: > > > > > > Hi Han, > > > > > > yes, I believe you are totally right. But it still feels like a chicken > > > and > > >

Re: [ovs-dev] [PATCH ovn v2] northd: support HW VTEP with stateful datapath

You’re absolutely right. The difference with normal chassis is only the lack of output port. It is determined in l2_lkp table. This is the only one reason to send packet to ingress pipeline. Regards, Vladislav Odintsov > On 18 Sep 2021, at 01:25, Numan Siddique wrote: > > On Fri, Sep 17, 2021

Re: [ovs-dev] [PATCH ovn v2] northd: support HW VTEP with stateful datapath

On Fri, Sep 17, 2021 at 5:59 PM Vladislav Odintsov wrote: > > Hi Numan, > > I’ve posted a new patch version here: > https://patchwork.ozlabs.org/project/ovn/patch/20210917215602.10633-1-odiv...@gmail.com/ Thanks. I'll take a look. If I understand correctly, packets coming from RAMP

Re: [ovs-dev] [PATCH ovn v3 3/3] ic: add support for routing tables in adv/learn routes

Bleep bloop. Greetings Vladislav Odintsov, 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 82 characters long (recommended limit is 79) #356 FILE:

Re: [ovs-dev] [PATCH ovn v3 2/3] northd, utils: support for RouteTables in LRs

Bleep bloop. Greetings Vladislav Odintsov, 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 bracing around statement #243 FILE: northd/northd.c:8878: else

Re: [ovs-dev] [PATCH ovn v3 1/3] ic: process only local port_bindings

Bleep bloop. Greetings Vladislav Odintsov, 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 80 characters long (recommended limit is 79) #48 FILE:

Re: [ovs-dev] [PATCH ovn v2] northd: support HW VTEP with stateful datapath

Hi Numan, I’ve posted a new patch version here: https://patchwork.ozlabs.org/project/ovn/patch/20210917215602.10633-1-odiv...@gmail.com/ I’ve tried to answer your question about ACLs in documentation. Please let me know if it is clear. Regards, Vladislav Odintsov > On 17 Sep 2021, at 22:42,

[ovs-dev] [PATCH ovn] northd: support HW VTEP with stateful datapath

A packet going from HW VTEP device to VIF port when arrives to hypervisor chassis should go through LS ingress pipeline to l2_lkp stage without any match. In l2_lkp stage an output port is determined and then packet passed to LS egress pipeline for futher processing and to VIF port delivery.

Re: [ovs-dev] [PATCH ovn 2/2] Update the probe interval in main loop.

On Fri, Sep 17, 2021 at 1:41 PM Han Zhou wrote: > > > > On Thu, Sep 16, 2021 at 8:06 PM Zhen Wang wrote: > > > > From: zhen wang > > > > When ovn-northd work in HA mode, ovn-northd will not update the probe > > interval in standby mode. This patch address the problem by updating > > the value

[ovs-dev] [PATCH ovn v3 3/3] ic: add support for routing tables in adv/learn routes

Previously support for multiple routing tables was added to northd code. This commit expands support for multiple routing tables by adding support of advertising and learning routes with their routing table information. To utilize such feature, user must: 1. create Logical Router in each AZ; 2.

[ovs-dev] [PATCH ovn v3 2/3] northd, utils: support for RouteTables in LRs

This patch extends Logical Router's routing functionality. Now user may create multiple routing tables within a Logical Router and assign them to Logical Router Ports. Traffic coming from Logical Router Port with assigned route_table is checked against global routes if any

[ovs-dev] [PATCH ovn v3 1/3] ic: process only local port_bindings

This commit adds a small optimization by utilizing ovsdb_index to iterate over port_bindings. Prior to this change each iteration checked availability_zone and continued processing only if port_binding belons to local AZ. Now we run against port_bindings from local AZ only and don't check

[ovs-dev] [PATCH ovn v3 0/3] Add multiple routing tables support to Logical Routers

This patch series extends Logical Router's routing functionality. Now user may create multiple routing tables within a Logical Router and assign them to Logical Router Ports. Traffic coming from Logical Router Port with assigned route_table is checked against global routes if any

Re: [ovs-dev] [PATCH ovn 1/2] Revert "northd: Don't poll ovsdb before the connection is fully established"

On Fri, Sep 17, 2021 at 2:04 PM Numan Siddique wrote: > > On Fri, Sep 17, 2021 at 5:02 PM Renat Nurgaliyev wrote: > > > > Hi Han, > > > > yes, I believe you are totally right. But it still feels like a chicken and > > egg problem to me, storing the database timeout setting inside the database >

Re: [ovs-dev] [PATCH ovn 1/2] Revert "northd: Don't poll ovsdb before the connection is fully established"

Hi Numan, according to our experience, in such case the connection can never be fully established, and northd ends up being in an endless loop with 100% CPU utilization. Numan Siddique schrieb am Sa., 18. Sept. 2021, 00:05: > On Fri, Sep 17, 2021 at 5:02 PM Renat Nurgaliyev > wrote: > > > >

Re: [ovs-dev] [PATCH ovn 1/2] Revert "northd: Don't poll ovsdb before the connection is fully established"

On Fri, Sep 17, 2021 at 5:02 PM Renat Nurgaliyev wrote: > > Hi Han, > > yes, I believe you are totally right. But it still feels like a chicken and > egg problem to me, storing the database timeout setting inside the database > itself. If there would be at least some local command line argument

Re: [ovs-dev] [PATCH ovn 1/2] Revert "northd: Don't poll ovsdb before the connection is fully established"

Hi Han, yes, I believe you are totally right. But it still feels like a chicken and egg problem to me, storing the database timeout setting inside the database itself. If there would be at least some local command line argument to override timeout value, it would be already amazing, because

Re: [ovs-dev] [PATCH ovn 1/2] Revert "northd: Don't poll ovsdb before the connection is fully established"

On Fri, Sep 17, 2021 at 1:48 PM Renat Nurgaliyev wrote: > > Hello Han, > > when I wrote this patch we had an issue with a very big SB database, around 1,5 gigabytes. There were no controllers or northds running, so the database server was without any load at all. Although OVSDB was idling, even a

Re: [ovs-dev] [PATCH ovn 1/2] Revert "northd: Don't poll ovsdb before the connection is fully established"

Hello Han, when I wrote this patch we had an issue with a very big SB database, around 1,5 gigabytes. There were no controllers or northds running, so the database server was without any load at all. Although OVSDB was idling, even a single northd process could not fully connect to the database

Re: [ovs-dev] [PATCH ovn v2 1/2] Prepare for 21.09.0.

I pushed patch 1 to master, then created the branch-21.09 branch, and pushed patch 2 to master. On 9/17/21 3:42 PM, Numan Siddique wrote: On Fri, Sep 17, 2021 at 2:44 PM Mark Michelson wrote: Signed-off-by: Mark Michelson For both the patches: Acked-by: Numan Siddique Numan ---

Re: [ovs-dev] [OVN Patch v8 1/3] northd: Disable parallel processing for logical_dp_groups

Based on the successful GHA run I had in my fork, I fixed the submodule downgrade (and fixed a couple of grammar errors in comments). I've pushed this to master. On 9/17/21 3:49 PM, Mark Michelson wrote: On 9/17/21 1:38 PM, Ilya Maximets wrote: On 9/15/21 14:43,

Re: [ovs-dev] [PATCH ovn 2/2] Update the probe interval in main loop.

On Thu, Sep 16, 2021 at 8:06 PM Zhen Wang wrote: > > From: zhen wang > > When ovn-northd work in HA mode, ovn-northd will not update the probe > interval in standby mode. This patch address the problem by updating > the value in main loop. > Thanks Zhen for the fix! Maybe the impact and steps

Re: [ovs-dev] [PATCH ovn 1/2] Revert "northd: Don't poll ovsdb before the connection is fully established"

On Fri, Sep 17, 2021 at 1:31 PM Numan Siddique wrote: > > On Thu, Sep 16, 2021 at 11:05 PM Zhen Wang via dev > wrote: > > > > From: zhen wang > > > > This reverts commit 1e59feea933610b28fd4442243162ce35595cfee. > > Above commit introduced a bug when muptiple ovn-northd instances work in HA > >

Re: [ovs-dev] [PATCH ovn] binding: Store timestamp at which ovn-installed was set.

On Wed, Sep 15, 2021 at 10:22 AM Dumitru Ceara wrote: > > CMSs (e.g., ovn-kubernetes) already use the ovn-installed > external_id set by ovn-controller in OVS.Interface.external_ids to > determine that the networking for a VIF is completely plugged. > > ovn-controller now also stores the

Re: [ovs-dev] [PATCH ovn] ovn.at: Skip tests that need tcpdump when it's not installed.

On Wed, Sep 15, 2021 at 10:26 AM Dumitru Ceara wrote: > > Signed-off-by: Dumitru Ceara Thanks. Applied. Numan > --- > tests/ovn.at | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/tests/ovn.at b/tests/ovn.at > index 30625ec37..dc6569008 100644 > --- a/tests/ovn.at > +++

Re: [ovs-dev] [PATCH ovn 1/2] Revert "northd: Don't poll ovsdb before the connection is fully established"

On Thu, Sep 16, 2021 at 8:05 PM Zhen Wang wrote: > > From: zhen wang > > This reverts commit 1e59feea933610b28fd4442243162ce35595cfee. > Above commit introduced a bug when muptiple ovn-northd instances work in HA > mode. If SB leader and active ovn-northd instance got killed by system power >

Re: [ovs-dev] [PATCH ovn 1/2] Revert "northd: Don't poll ovsdb before the connection is fully established"

On Thu, Sep 16, 2021 at 11:05 PM Zhen Wang via dev wrote: > > From: zhen wang > > This reverts commit 1e59feea933610b28fd4442243162ce35595cfee. > Above commit introduced a bug when muptiple ovn-northd instances work in HA > mode. If SB leader and active ovn-northd instance got killed by system

Re: [ovs-dev] [PATCH ovn] rhel, utils: don't affect traffic on controller upgrade

On Fri, Sep 17, 2021 at 9:28 AM Vladislav Odintsov wrote: > > Currently upgrade of ovn-host rpm package affects active > traffic. This is because systemctl try-restart > ovn-controller is invoked during rpm package upgrade. > It calls ovn-ctl stop_controller and then start_controller. > > Adding

Re: [ovs-dev] [OVN Patch v8 1/3] northd: Disable parallel processing for logical_dp_groups

On 9/17/21 1:38 PM, Ilya Maximets wrote: On 9/15/21 14:43, anton.iva...@cambridgegreys.com wrote: From: Anton Ivanov Work on improving processing with dp_groups enabled has discovered that the locking mechanism presently in use is not reliable. Disabling parallel processing if dp_groups are

Re: [ovs-dev] [PATCH ovn v2 1/2] Prepare for 21.09.0.

On Fri, Sep 17, 2021 at 2:44 PM Mark Michelson wrote: > > Signed-off-by: Mark Michelson For both the patches: Acked-by: Numan Siddique Numan > --- > NEWS | 4 ++-- > configure.ac | 2 +- > debian/changelog | 4 ++-- > 3 files changed, 5 insertions(+), 5 deletions(-) > >

Re: [ovs-dev] [PATCH ovn v2] northd: support HW VTEP with stateful datapath

On Fri, Sep 17, 2021 at 11:01 AM Vladislav Odintsov wrote: > > A packet going from HW VTEP device to VIF port when arrives to > hypervisor chassis should go through LS ingress pipeline to l2_lkp > stage without any match. In l2_lkp stage an output port is > determined and then packet passed to LS

Re: [ovs-dev] [PATCH v6 ovn] controller: add datapath meter capability check

I merged this to master. Thanks Lorenzo and Mark! On 9/17/21 12:25 PM, Lorenzo Bianconi wrote: Dump datapath meter capabilities before configuring meters in ovn-controller Signed-off-by: Lorenzo Bianconi --- Changes since v5: - send new command only on reconnection Changes since v4: - move

Re: [ovs-dev] [PATCH ovn v4 8/9] binding: Consider plugging of ports on CMS request.

On Fri, Sep 17, 2021 at 12:41 PM Frode Nordahl wrote: > > On Fri, Sep 17, 2021 at 1:16 AM Numan Siddique wrote: > > > > On Fri, Sep 3, 2021 at 3:28 PM Frode Nordahl > > wrote: > > > > > > When OVN is linked with an appropriate plugging implementation, > > > CMS can request OVN to plug

[ovs-dev] [PATCH ovn v2 2/2] Prepare for post-21.09.0.

Signed-off-by: Mark Michelson --- NEWS | 3 +++ configure.ac | 2 +- debian/changelog | 6 ++ 3 files changed, 10 insertions(+), 1 deletion(-) diff --git a/NEWS b/NEWS index 784240fbc..8a21c029e 100644 --- a/NEWS +++ b/NEWS @@ -1,3 +1,6 @@ +Post v21.09.0 +- +

[ovs-dev] [PATCH ovn v2 1/2] Prepare for 21.09.0.

Signed-off-by: Mark Michelson --- NEWS | 4 ++-- configure.ac | 2 +- debian/changelog | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/NEWS b/NEWS index 6e81f7bee..784240fbc 100644 --- a/NEWS +++ b/NEWS @@ -1,5 +1,5 @@ -Post-v21.06.0

Re: [ovs-dev] [OVN Patch v8 1/3] northd: Disable parallel processing for logical_dp_groups

On 17/09/2021 18:23, Mark Michelson wrote: Hi Anton, For the series: Acked-by: Mark Michelson Thanks. On 9/15/21 8:43 AM, anton.iva...@cambridgegreys.com wrote: From: Anton Ivanov Work on improving processing with dp_groups enabled has discovered that the locking mechanism presently in

Re: [ovs-dev] [PATCH v2] openvswitch: Fix condition check in do_execute_actions() by using nla_ok()

On Fri, 17 Sep 2021 08:07:14 + Jiasheng Jiang wrote: > Just using 'rem > 0' might be unsafe, so it's better > to use the nla_ok() instead. > Because we can see from the nla_next() that > '*remaining' might be smaller than 'totlen'. And nla_ok() > will avoid it happening. > For example,

Re: [ovs-dev] [OVN Patch v8 1/3] northd: Disable parallel processing for logical_dp_groups

On 9/15/21 14:43, anton.iva...@cambridgegreys.com wrote: > From: Anton Ivanov > > Work on improving processing with dp_groups enabled has > discovered that the locking mechanism presently in use > is not reliable. Disabling parallel processing if dp_groups > are enabled until the root cause is

Re: [ovs-dev] [PATCH ovn 1/2] Prepare for 21.09.0.

On 9/17/21 1:15 PM, Ilya Maximets wrote: On 9/17/21 18:13, Mark Michelson wrote: Signed-off-by: Mark Michelson --- NEWS | 4 ++-- configure.ac | 2 +- debian/changelog | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/NEWS b/NEWS index

Re: [ovs-dev] [PATCH v6 ovn] controller: add datapath meter capability check

On 17/09/2021 17:25, Lorenzo Bianconi wrote: > Dump datapath meter capabilities before configuring meters in > ovn-controller > > Signed-off-by: Lorenzo Bianconi > --- > Changes since v5: > - send new command only on reconnection > Changes since v4: > - move rconn setup in

Re: [ovs-dev] [OVN Patch v8 1/3] northd: Disable parallel processing for logical_dp_groups

Hi Anton, For the series: Acked-by: Mark Michelson On 9/15/21 8:43 AM, anton.iva...@cambridgegreys.com wrote: From: Anton Ivanov Work on improving processing with dp_groups enabled has discovered that the locking mechanism presently in use is not reliable. Disabling parallel processing if

Re: [ovs-dev] [PATCH] ovsdb-data: Optimize union of sets.

On 9/16/21 22:19, Ilya Maximets wrote: > Current algorithm of ovsdb_datum_union looks like this: > > for-each atom in b: > if not bin_search(a, atom): > a[n++] = clone(atom) > quicksort(a) > > So, the complexity looks like this: > >Nb * log2(Na) +Nb + (Na +

[ovs-dev] [PATCH v2 2/2] ovsdb-data: Optimize subtraction of sets.

Current algorithm for ovsdb_datum_subtract looks like this: for-each atom in a: if atom in b: swap(atom, ) destroy(atom) quicksort(a) Complexity: Na * log2(Nb) + (Na - Nb) * log2(Na - Nb) Search Comparisons for quicksort It's not optimal,

[ovs-dev] [PATCH v2 1/2] ovsdb-data: Optimize union of sets.

Current algorithm of ovsdb_datum_union looks like this: for-each atom in b: if not bin_search(a, atom): push(a, clone(atom)) quicksort(a) So, the complexity looks like this: Nb * log2(Na) +Nb + (Na + Nb) * log2(Na + Nb) Comparisonsclones

[ovs-dev] [PATCH v2 0/2] ovsdb: Optimize set operations.

Ilya Maximets (2): ovsdb-data: Optimize union of sets. ovsdb-data: Optimize subtraction of sets. lib/db-ctl-base.c | 10 ++-- lib/ovsdb-data.c | 136 +- lib/ovsdb-data.h | 6 +- lib/ovsdb-idl.c | 8 +-- ovsdb/mutation.c | 2 +-

Re: [ovs-dev] [PATCH ovn 1/2] Prepare for 21.09.0.

On 9/17/21 18:13, Mark Michelson wrote: > Signed-off-by: Mark Michelson > --- > NEWS | 4 ++-- > configure.ac | 2 +- > debian/changelog | 4 ++-- > 3 files changed, 5 insertions(+), 5 deletions(-) > > diff --git a/NEWS b/NEWS > index 6e81f7bee..c6671b8f5 100644 > --- a/NEWS >

Re: [ovs-dev] [PATCH ovn v4 4/9] controller: Move OVS port functions to new module.

On Fri, Sep 17, 2021 at 6:17 PM Mark Michelson wrote: > > Hi Frode, > > The changes I list below are all minor and could be taken care of by > whoever may merge this series. I don't think it's worth spinning a new > version of the series for these changes alone. However, I do see that > there

Re: [ovs-dev] [PATCH ovn v4 8/9] binding: Consider plugging of ports on CMS request.

On Fri, Sep 17, 2021 at 1:16 AM Numan Siddique wrote: > > On Fri, Sep 3, 2021 at 3:28 PM Frode Nordahl > wrote: > > > > When OVN is linked with an appropriate plugging implementation, > > CMS can request OVN to plug individual lports into the local > > Open vSwitch instance. > > > > The port and

[ovs-dev] [PATCH v6 ovn] controller: add datapath meter capability check

Dump datapath meter capabilities before configuring meters in ovn-controller Signed-off-by: Lorenzo Bianconi --- Changes since v5: - send new command only on reconnection Changes since v4: - move rconn setup in ovs_feature_support_update and rename it in ovs_feature_support_run - drop

Re: [ovs-dev] [PATCH ovn v4 4/9] controller: Move OVS port functions to new module.

Hi Frode, The changes I list below are all minor and could be taken care of by whoever may merge this series. I don't think it's worth spinning a new version of the series for these changes alone. However, I do see that there were other findings by Numan for later patches. On 9/3/21 3:27

Re: [ovs-dev] [PATCH v2] dpif-netdev: Fix pmd thread comments to include SMC.

On 09/09/2021 16:30, Cian Ferriter wrote: > These comments are relevant to SMC too. > > Fixes: 60d8ccae135f ("dpif-netdev: Add SMC cache after EMC cache") > Signed-off-by: Cian Ferriter > Acked-by: Kevin Traynor > --- Just to confirm, lgtm. Ilya, this is a minor non-functional change. If you

[ovs-dev] [PATCH ovn 1/2] Prepare for 21.09.0.

Signed-off-by: Mark Michelson --- NEWS | 4 ++-- configure.ac | 2 +- debian/changelog | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/NEWS b/NEWS index 6e81f7bee..c6671b8f5 100644 --- a/NEWS +++ b/NEWS @@ -1,5 +1,5 @@ -Post-v21.06.0

[ovs-dev] [PATCH ovn 2/2] Prepare for post-21.09.0.

Signed-off-by: Mark Michelson --- NEWS | 3 +++ configure.ac | 2 +- debian/changelog | 6 ++ 3 files changed, 10 insertions(+), 1 deletion(-) diff --git a/NEWS b/NEWS index c6671b8f5..dfd9b1f19 100644 --- a/NEWS +++ b/NEWS @@ -1,3 +1,6 @@ +Post v21.09.0 +- +

Re: [ovs-dev] [PATCH ovn 3/3] northd: support HW VTEP with stateful datapath

Hi, done: https://patchwork.ozlabs.org/project/ovn/patch/20210917150104.6143-1-odiv...@gmail.com/ Regards, Vladislav Odintsov > On 17 Sep 2021, at 17:44, Numan Siddique wrote: > > On Wed, Sep 15, 2021 at 8:07 PM Vladislav Odintsov > wrote: >> >> A packet going

[ovs-dev] [PATCH ovn v2] northd: support HW VTEP with stateful datapath

A packet going from HW VTEP device to VIF port when arrives to hypervisor chassis should go through LS ingress pipeline to l2_lkp stage without any match. In l2_lkp stage an output port is determined and then packet passed to LS egress pipeline for futher processing and to VIF port delivery.

Re: [ovs-dev] [PATCH ovn 3/3] northd: support HW VTEP with stateful datapath

On Wed, Sep 15, 2021 at 8:07 PM Vladislav Odintsov wrote: > > A packet going from HW VTEP device to VIF port when arrives to > hypervisor chassis should go through LS ingress pipeline to l2_lkp > stage without any match. In l2_lkp stage an output port is > determined and then packet passed to LS

Re: [ovs-dev] [PATCH ovn 2/3] tests: check that flow for ramp switch is added when PB is created

On Wed, Sep 15, 2021 at 8:07 PM Vladislav Odintsov wrote: > > Signed-off-by: Vladislav Odintsov Thanks. I applied the first 2 patches of this series to the main branch. Patch 3 needs a rebase. Numan > --- > tests/ovn-controller-vtep.at | 64 > 1 file

Re: [ovs-dev] [PATCH ovn] controller: Fall back to full recompute of pflow_output for vtep lport changes.

On Wed, Sep 15, 2021 at 8:12 PM Odintsov Vladislav wrote: > > Hi Numan, > > for me this patch solves the issue. > Actually, it’s a good point about adding test for ovn-controller-vtep to > eliminate this problem in future. > However there was another problem with HW VTEP support in OVN with

[ovs-dev] [PATCH ovn] rhel, utils: don't affect traffic on controller upgrade

Currently upgrade of ovn-host rpm package affects active traffic. This is because systemctl try-restart ovn-controller is invoked during rpm package upgrade. It calls ovn-ctl stop_controller and then start_controller. Adding ovn-ctl stop_controller --restart to %postun upgrade case right before

[ovs-dev] dpif-netdev.c : remove dpif OVS_UNUSED flag.

From: linhuang The dpif is used, so remove the OVS_UNUSED flag. Signed-off-by: linhuang --- lib/dpif-netdev.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c index 4cd0e9ebd..75f381ec1 100644 --- a/lib/dpif-netdev.c +++

Re: [ovs-dev] [PATCH v5 ovn] controller: add datapath meter capability check

On 17/09/2021 11:12, Lorenzo Bianconi wrote: > Dump datapath meter capabilities before configuring meters in > ovn-controller > > Signed-off-by: Lorenzo Bianconi > --- > Changes since v4: > - move rconn setup in ovs_feature_support_update and rename it in > ovs_feature_support_run > - drop

[ovs-dev] [PATCH v5 ovn] controller: add datapath meter capability check

Dump datapath meter capabilities before configuring meters in ovn-controller Signed-off-by: Lorenzo Bianconi --- Changes since v4: - move rconn setup in ovs_feature_support_update and rename it in ovs_feature_support_run - drop ovs_feature_support_init() - rename ovs_feature_support_deinit in

Re: [ovs-dev] [PATCH 0/4] dpif-netdev: rxq auto-lb improvements

Hi Anurag, On 17/09/2021 06:50, Anurag Agarwal wrote: > Thanks Ilya for your email. Except for one patch, all the others are > superseded by the patchset posted by Kevin. > > Outstanding patch that should be applicable is the following one: > > [ovs-dev] [PATCH 4/4] dpif-netdev: Allow

[ovs-dev] [PATCH 2/2] openvswitch: Fix condition check in output_userspace() by using nla_ok()

Just using 'rem > 0' might be unsafe, so it's better to use the nla_ok() instead. Because we can see from the nla_next() that '*remaining' might be smaller than 'totlen'. And nla_ok() will avoid it happening. For example, ovs_dp_process_packet() -> ovs_execute_actions() -> do_execute_actions() ->

Re: [ovs-dev] [PATCH v4 ovn] controller: add datapath meter capability check

On 16/09/2021 23:16, Lorenzo Bianconi wrote: > Dump datapath meter capabilities before configuring meters in > ovn-controller Thanks. Generally OK with this with one suggestion below. I'm not sure if it will work but I think it follows the typical ovn/ovn pattern which uses _run() functions in

[ovs-dev] [PATCH v2] openvswitch: Fix condition check in do_execute_actions() by using nla_ok()

Just using 'rem > 0' might be unsafe, so it's better to use the nla_ok() instead. Because we can see from the nla_next() that '*remaining' might be smaller than 'totlen'. And nla_ok() will avoid it happening. For example, ovs_dp_process_packet() -> ovs_execute_actions() -> do_execute_actions(),

Re: [ovs-dev] [PATCH ovn v5 0/2] northd: Split northd

On 16/09/2021 21:47, Numan Siddique wrote: > On Thu, Sep 16, 2021 at 3:06 PM Han Zhou wrote: >> >> Acked-by: Han Zhou > > Thanks Mark G (and Mark M and Han). I applied both the patches to the > main branch. Thanks! > > Numan > >> >> On Thu, Sep 16, 2021 at 8:49 AM Mark Gray wrote: >> >>>