With the help of logical_flow's in_out_port tag, we can skip parsing a
big portion of the logical flows in SB DB, which can largely improve
ovn-controller's performance whenever a full recompute is required.
With a scale test topology of 1000 chassises, 20 LSPs per chassis, 20k
lports in total
Populate the in_out_port tag for logical switch pipeline flows wherever
possible.
Signed-off-by: Han Zhou
---
northd/ovn-northd.c | 272 +---
1 file changed, 155 insertions(+), 117 deletions(-)
diff --git a/northd/ovn-northd.c b/northd/ovn-northd.c
index
The column will provide information to help improve efficiency of
ovn-controller lflow parsing.
Signed-off-by: Han Zhou
---
ovn-sb.ovsschema | 7 +--
ovn-sb.xml | 23 +++
2 files changed, 28 insertions(+), 2 deletions(-)
diff --git a/ovn-sb.ovsschema
With the help of a new column in Logical_Flow table that stores ingress/egress
lport information, ovn-controller can avoid parsing a big portion of the
logical flows in SB DB, which can largely improve ovn-controller's performance
whenever a full recompute is required.
With a scale test topology
When counting lsp specific flows, using format "table" for ovn-sbctl
output to make sure each record is counted at most once.
Signed-off-by: Han Zhou
---
tests/ovn-northd.at | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/ovn-northd.at b/tests/ovn-northd.at
index
On Wed, Jun 30, 2021 at 4:48 PM Rosemarie O'Riorden wrote:
>
> From: Rosemarie O'Riorden
>
> This change removes the default values for EAL args socket-mem and
> socket-limit. As DPDK supports dynamic memory allocation, there is no
> need to allocate a certain amount of memory on start-up, nor
On Wed, Jun 30, 2021 at 4:48 PM Rosemarie O'Riorden wrote:
>
> From: Rosemarie O'Riorden
>
> This change removes the automatic memory limit on start-up of OVS with
> DPDK. As DPDK supports dynamic memory allocation, there is no
> need to limit the amount of memory available, if not requested.
>
Bleep bloop. Greetings Mark Michelson, 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 84 characters long (recommended limit is 79)
#162 FILE:
Bleep bloop. Greetings Mark Michelson, 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
#92 FILE: northd/ovn-northd.c:1493:
if
Previously, ARP TPAs were filtered down only to "reachable" addresses.
Reachable addresses are all router interface addresses, as well as NAT
external addresses and load balancer VIPs that are within the subnet
handled by a router's port.
However, it is possible that in some configurations, CMSes
Dealing with NAT and load balancer IPs has been a bit of a pain point.
It requires creating static routes if east-west traffic to those
addresses is desired. Further, it requires ARPs to be sent between the
logical routers in order to create MAC Bindings.
This commit seeks to make things easier.
Load_Balancer and NAT entries have a new option, "add_route" that can be
set to automatically add routes to those addresses to neighbor routers,
therefore eliminating the need to create static routes.
Signed-off-by: Mark Michelson
---
northd/ovn-northd.8.xml | 7 -
northd/ovn-northd.c
The same pattern is repeated several times throughout ovn-northd.c, so
this puts it in its own function. This will be used even more in an
upcoming commit.
Signed-off-by: Mark Michelson
---
northd/ovn-northd.c | 70 -
1 file changed, 24 insertions(+),
This patch series aims to fix issues seen in OpenStack deployments when
floating IPs were assigned to routers, and those floating IPs were not
part of any subnet configured on that router.
This series has gone through several transformations, but this
incarnation is a four patch series:
Patch 1
On Wed, Jun 30, 2021 at 7:01 AM Dumitru Ceara wrote:
>
> It's valid that port_groups contain non-vif ports, they can actually
> contain any type of logical_switch_port.
>
> Also, there's no need to allocate a new sset containing the local ports'
> names every time the I-P engine processes a
On Tue, Jun 22, 2021 at 1:17 AM Han Zhou wrote:
>
> Acked-by: Han Zhou
Thanks, I applied this patch.
Numan
>
> On Sat, Jun 19, 2021 at 2:51 AM Mark Gray wrote:
>
> > Signed-off-by: Mark Gray
> > ---
> > tests/system-ovn.at | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> >
From: Rosemarie O'Riorden
This change removes the automatic memory limit on start-up of OVS with
DPDK. As DPDK supports dynamic memory allocation, there is no
need to limit the amount of memory available, if not requested.
Currently, if socket-limit is not configured, it is set to the value of
From: Rosemarie O'Riorden
This change removes the default values for EAL args socket-mem and
socket-limit. As DPDK supports dynamic memory allocation, there is no
need to allocate a certain amount of memory on start-up, nor limit the
amount of memory available, if not requested.
Currently,
From: Rosemarie O'Riorden
Currently, there is a default value of 1024 for socket-mem if not
configured. socket-limit automatically takes on the value of socket-mem
unless otherwise specified. With these changes, memory allocation will
be dynamically managed by DPDK, meaning that by default, no
Hi,
On Thu, Jun 17, 2021 at 09:57:53PM +0530, Kumar Amber wrote:
> From: Harry van Haaren
>
> This commit adds 3 new traffic profile implementations to the
> existing avx512 miniflow extract infrastructure. The profiles added are:
> - Ether()/IP()/TCP()
> - Ether()/Dot1Q()/IP()/UDP()
> -
Hi Mark,
Thank you for the patch! Perhaps something to improve:
[auto build test WARNING on net-next/master]
url:
https://github.com/0day-ci/linux/commits/Mark-Gray/openvswitch-Introduce-per-cpu-upcall-dispatch/20210630-175435
base: https://git.kernel.org/pub/scm/linux/kernel/git/davem
On Thu, Jun 3, 2021 at 2:50 PM Mark Michelson wrote:
>
> This patch series aims to fix issues seen in OpenStack deployments when
> floating IPs were assigned to routers, and those floating IPs were not
> part of any subnet configured on that router.
>
> Originally, this was a two patch series but
> -Original Message-
> From: Amber, Kumar
> Sent: Wednesday, June 30, 2021 4:10 PM
> To: Eelco Chaudron ; Van Haaren, Harry
>
> Cc: d...@openvswitch.org; i.maxim...@ovn.org; Flavio Leitner
> ;
> Stokes, Ian
> Subject: RE: [ovs-dev] [v4 11/12] dpif-netdev/mfex: add more AVX512 traffic
> -Original Message-
> From: Eelco Chaudron
> Sent: Wednesday, June 30, 2021 3:35 PM
> To: Van Haaren, Harry
> Cc: Amber, Kumar ; d...@openvswitch.org;
> i.maxim...@ovn.org; Flavio Leitner ; Stokes, Ian
>
> Subject: Re: [ovs-dev] [v4 10/12] dpif-netdev/mfex: Add AVX512 based optimized
>
Hi Guru,
Thank you for letting me know. I backported the change until 2.13.
Thank you,
Alin.
From: Guru Shetty
Sent: Saturday, June 26, 2021 1:14 AM
To: Alin-Gabriel Serdean
Cc: Ilya Maximets ; ovs dev ; Frank
Wagner
Subject: Re: [ovs-dev] [PATCH] datapath-windows: Specify
The match keyword "igmp" is not supported in ofp-parse, which means
that flow dumps cannot be restored. This patch prints the igmp match
in the accepted format (ip,nw_proto=2) and adds a test.
Signed-off-by: Adrian Moreno
---
lib/match.c| 2 --
tests/ovs-ofctl.at | 2 ++
2 files
As discussed in [1], "igmp" is not a special OF header so
ofp_parse_protocol() does not accept it. However, it's still being
printed by match_format().
This makes flow dump restoration fail if it contains this match,
which affects for instance the ovs-save script used by "ovs-ctl restart"
to make
Thanks Eelco sure go ahead :) upgrade will surely raise the pps!
> -Original Message-
> From: Eelco Chaudron
> Sent: Wednesday, June 30, 2021 8:27 PM
> To: Amber, Kumar ; Van Haaren, Harry
>
> Cc: d...@openvswitch.org; i.maxim...@ovn.org; Stokes, Ian
> ; Flavio Leitner
> Subject: Re:
Hi Eelco,
Pls find my comments inline.
> > #define KMASK_ETHER 0x1FFFULL
> > +#define KMASK_DT1Q 0x000FULL
>
> This was messing me up, as this suggests this is a 16-byte mask, but this is
> only 8,
> so maybe we should indicate it by removing the two leading zeros?
>
>#define
No additional comments on this patch! This concludes my review of v4, looking
forward to v5.
I will now do some additional tests on my non AVX512 machine. Guess I need to
update my Intel NUC to an AVX512 supported one :)
//Eelco
On 17 Jun 2021, at 18:27, Kumar Amber wrote:
> From: Harry van
On 17 Jun 2021, at 18:27, Kumar Amber wrote:
> From: Harry van Haaren
>
> This commit adds 3 new traffic profile implementations to the
> existing avx512 miniflow extract infrastructure. The profiles added are:
> - Ether()/IP()/TCP()
> - Ether()/Dot1Q()/IP()/UDP()
> -
On 30 Jun 2021, at 15:30, Van Haaren, Harry wrote:
>> -Original Message-
>> From: Eelco Chaudron
>> Sent: Wednesday, June 30, 2021 2:12 PM
>> To: Amber, Kumar ; Van Haaren, Harry
>>
>> Cc: d...@openvswitch.org; i.maxim...@ovn.org; Flavio Leitner
>> ;
>> Stokes, Ian
>> Subject: Re:
On 6/28/21 8:05 AM, Han Zhou wrote:
> On Fri, Jun 25, 2021 at 2:38 PM Numan Siddique wrote:
>>
>> On Fri, Jun 25, 2021 at 2:53 PM Han Zhou wrote:
>>>
>>> On Fri, Jun 25, 2021 at 4:50 AM Dumitru Ceara wrote:
It's valid that port_groups contain non-vif ports, they can actually
It's valid that port_groups contain non-vif ports, they can actually
contain any type of logical_switch_port.
Also, there's no need to allocate a new sset containing the local ports'
names every time the I-P engine processes a change. We were already
maintaining a set of "local_lport_ids".
On 6/30/21 2:49 PM, Mark Gray wrote:
> On 24/06/2021 16:33, Dumitru Ceara wrote:
>> On 6/18/21 10:52 AM, Mark Gray wrote:
>>> For performance measurement, it is useful to understand the
>>> length of time required to complete a number of key code paths
>>> in ovn-northd.c. Add stopwatches to
On 6/30/21 2:49 PM, Mark Gray wrote:
> On 24/06/2021 16:34, Dumitru Ceara wrote:
>> On 6/18/21 10:52 AM, Mark Gray wrote:
>>> Add a suite of micro-benchmarks to aid a developer in understanding the
>>> performance impact of any changes that they are making. They can be used to
>>> help to
On Tue, Jun 29, 2021 at 1:18 PM Ben Pfaff wrote:
>
> On Tue, Jun 29, 2021 at 12:50:54PM -0400, Ihar Hrachyshka wrote:
> > On Tue, Jun 29, 2021 at 12:46 PM Ben Pfaff wrote:
> > >
> > > On Tue, Jun 29, 2021 at 12:24:11PM -0400, Ihar Hrachyshka wrote:
> > > > One, HEAD~, not HEAD, should be used to
> -Original Message-
> From: Eelco Chaudron
> Sent: Wednesday, June 30, 2021 11:07 AM
> To: Van Haaren, Harry
> Cc: Flavio Leitner ; Amber, Kumar ;
> d...@openvswitch.org; i.maxim...@ovn.org
> Subject: Re: [ovs-dev] [v4 03/12] dpif-netdev: Add study function to select
> the best
> mfex
> -Original Message-
> From: Eelco Chaudron
> Sent: Wednesday, June 30, 2021 2:12 PM
> To: Amber, Kumar ; Van Haaren, Harry
>
> Cc: d...@openvswitch.org; i.maxim...@ovn.org; Flavio Leitner
> ;
> Stokes, Ian
> Subject: Re: [ovs-dev] [v4 10/12] dpif-netdev/mfex: Add AVX512 based
Add a suite of micro-benchmarks to aid a developer in understanding the
performance impact of any changes that they are making. They can be used to
help to understand the relative performance between two test runs on the same
test machine, but are not intended to give the absolute performance of
This is a proposal to add some micro-benchmarks to aid developers
in benchmarking optimizations to OVN. It starts by adding simple
metrics for northd but could be expanded in future patches.
Mark Gray (2):
ovn-northd: Add useful stopwatches
tests: Add check-perf target
For performance measurement, it is useful to understand the
length of time required to complete a number of key code paths
in ovn-northd.c. Add stopwatches to measure these timings.
Signed-off-by: Mark Gray
Acked-by: Dumitru Ceara
---
Notes:
v4: Add common header file for stopwatch names
This patch was an interesting patch to review and being reminded about
endianness, and this site,
https://software.intel.com/sites/landingpage/IntrinsicsGuide/#text=_mm512_maskz_permutexvar_epi8=4315,
got me through it ;)
Some comments below...
//Eelco
On 17 Jun 2021, at 18:27, Kumar Amber
On Wed, Jun 30, 2021, at 11:09, Ilya Maximets wrote:
> On 5/20/21 3:35 PM, Gaetan Rivet wrote:
> > Joining pthreads makes the caller quiescent. It should register as such,
> > as joined threads may wait on an RCU callback executing before quitting,
> > deadlocking the caller.
>
> Hi, Gaetan.
>
>
Bleep bloop. Greetings Mark Gray, 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.
build:
libtool: link: gcc -std=gnu99 -Wstrict-prototypes -Wall -Wextra
-Wno-sign-compare -Wpointer-arith
Add a suite of micro-benchmarks to aid a developer in understanding the
performance impact of any changes that they are making. They can be used to
help to understand the relative performance between two test runs on the same
test machine, but are not intended to give the absolute performance of
For performance measurement, it is useful to understand the
length of time required to complete a number of key code paths
in ovn-northd.c. Add stopwatches to measure these timings.
Signed-off-by: Mark Gray
Acked-by: Dumitru Ceara
---
Notes:
v4: Add common header file for stopwatch names
This is a proposal to add some micro-benchmarks to aid developers
in benchmarking optimizations to OVN. It starts by adding simple
metrics for northd but could be expanded in future patches.
Mark Gray (2):
ovn-northd: Add useful stopwatches
tests: Add check-perf target
On 24/06/2021 16:34, Dumitru Ceara wrote:
> On 6/18/21 10:52 AM, Mark Gray wrote:
>> Add a suite of micro-benchmarks to aid a developer in understanding the
>> performance impact of any changes that they are making. They can be used to
>> help to understand the relative performance between two
On 24/06/2021 16:33, Dumitru Ceara wrote:
> On 6/18/21 10:52 AM, Mark Gray wrote:
>> For performance measurement, it is useful to understand the
>> length of time required to complete a number of key code paths
>> in ovn-northd.c. Add stopwatches to measure these timings.
>>
>> Signed-off-by: Mark
> On Fri, Jun 18, 2021 at 9:04 AM Lorenzo Bianconi
> wrote:
> >
> > Rework lb flow logic in order to visit first each load_balancer and then
> > related datapath during lb flow installation.
> > This patch allows to reduce memory footprint and cpu utilization in
> > ovn-northd.
> >
>
> Hi
> -Original Message-
> From: Eelco Chaudron
> Sent: Wednesday, June 30, 2021 10:52 AM
> To: Van Haaren, Harry
> Cc: Amber, Kumar ; d...@openvswitch.org;
> i.maxim...@ovn.org
> Subject: Re: [ovs-dev] [v4 03/12] dpif-netdev: Add study function to select
> the best
> mfex function
>
>
>
On 30 Jun 2021, at 11:43, Van Haaren, Harry wrote:
>> -Original Message-
>> From: Flavio Leitner
>> Sent: Tuesday, June 29, 2021 7:11 PM
>> To: Van Haaren, Harry
>> Cc: Eelco Chaudron ; Amber, Kumar
>> ; d...@openvswitch.org; i.maxim...@ovn.org
>> Subject: Re: [ovs-dev] [v4 03/12]
On 30/06/2021 10:56, Mark Gray wrote:
> This series proposes a new method of distributing upcalls
> to user space threads attempting to resolve a number of
> issues with the current method.
>
> Mark Gray (3):
> ofproto: change type of n_handlers and n_revalidators
> dpif-netlink: fix
The Open vSwitch kernel module uses the upcall mechanism to send
packets from kernel space to user space when it misses in the kernel
space flow table. The upcall sends packets via a Netlink socket.
Currently, a Netlink socket is created for every vport. In this way,
there is a 1:1 mapping between
Fixes: 1579cf677fcb ("dpif-linux: Implement the API functions to allow multiple
handler threads read upcall.")
Signed-off-by: Mark Gray
---
Notes:
v1 - Reworked based on Flavio's comments:
* Added "Fixes" tag
lib/dpif-netlink.c | 4 ++--
1 file changed, 2 insertions(+), 2
This series proposes a new method of distributing upcalls
to user space threads attempting to resolve a number of
issues with the current method.
Mark Gray (3):
ofproto: change type of n_handlers and n_revalidators
dpif-netlink: fix report_loss() message
dpif-netlink: Introduce per-cpu
'n_handlers' and 'n_revalidators' are declared as type 'size_t'.
However, dpif_handlers_set() requires parameter 'n_handlers' as
type 'uint32_t'. This patch fixes this type mismatch.
Signed-off-by: Mark Gray
---
Notes:
v1 - Reworked based on Flavio's comments:
* fixed inconsistency
Hi Eelco ,
Replies inline.
> -Original Message-
> From: Eelco Chaudron
> Sent: Wednesday, June 30, 2021 3:17 PM
> To: Amber, Kumar ; Van Haaren, Harry
>
> Cc: d...@openvswitch.org; i.maxim...@ovn.org; Flavio Leitner
>
> Subject: Re: [ovs-dev] [v4 01/12] dpif-netdev: Add command line
The Open vSwitch kernel module uses the upcall mechanism to send
packets from kernel space to user space when it misses in the kernel
space flow table. The upcall sends packets via a Netlink socket.
Currently, a Netlink socket is created for every vport. In this way,
there is a 1:1 mapping between
On 30 Jun 2021, at 11:32, Van Haaren, Harry wrote:
>> -Original Message-
>> From: Eelco Chaudron
>> Sent: Wednesday, June 30, 2021 10:18 AM
>> To: Van Haaren, Harry
>> Cc: Amber, Kumar ; d...@openvswitch.org;
>> i.maxim...@ovn.org
>> Subject: Re: [ovs-dev] [v4 03/12] dpif-netdev: Add
On 17 Jun 2021, at 18:27, Kumar Amber wrote:
> This patch introduces the mfex function pointers which allows
> the user to switch between different miniflow extract implementations
> which are provided by the OVS based on optimized ISA CPU.
>
> The user can query for the available minflow
On 19/05/2021 22:47, Pravin Shelar wrote:
Thanks for the review and sorry for the delay. I will be more responsive
to any further changes from this point on.
> On Fri, Apr 30, 2021 at 8:33 AM Mark Gray wrote:
>>
>> The Open vSwitch kernel module uses the upcall mechanism to send
>> packets from
On 28/05/2021 20:50, Flavio Leitner wrote:
> On Fri, Apr 30, 2021 at 11:31:27AM -0400, Mark Gray wrote:
>> 'n_handlers' and 'n_revalidators' are declared as type 'size_t'.
>> However, dpif_handlers_set() requires parameter 'n_handlers' as
>> type 'uint32_t'. This patch fixes this type mismatch.
>
On 28/05/2021 20:49, Flavio Leitner wrote:
> On Fri, Apr 30, 2021 at 11:31:28AM -0400, Mark Gray wrote:
>> Signed-off-by: Mark Gray
> This looks like a bug fix for this commit:
> 1579cf677fcb dpif-linux: Implement the API functions to allow multiple ...
>
> If you agree, please add the Fixes:
On 28/05/2021 20:49, Flavio Leitner wrote:
>
> Hi Mark,
>
> I think this patch is going in the right direction but there
> are some points that I think we should address. See below.
>
> On Fri, Apr 30, 2021 at 11:33:25AM -0400, Mark Gray wrote:
>> The Open vSwitch kernel module uses the upcall
On 08/06/2021 21:07, Flavio Leitner wrote:
>
> Hi Mark,
>
> This looks good to me.
>
> Since the new scheme doesn't allow users to change the number
> of handlers, we must update ovs-vswitchd.conf.db(5) as well.
I updated this documentation
>
> Some comments below.
>
> On Fri, Apr 30, 2021
On 12/05/2021 03:11, Flavio Leitner wrote:
> The results show that this new patch-set addressed the main
> thundering herd issue and the scalability issue I reported
> during V10 review.
>
> Unfortunately I can review the patches only next week.
Thanks again for testing and reviewing. Sorry for
> -Original Message-
> From: Flavio Leitner
> Sent: Tuesday, June 29, 2021 7:11 PM
> To: Van Haaren, Harry
> Cc: Eelco Chaudron ; Amber, Kumar
> ; d...@openvswitch.org; i.maxim...@ovn.org
> Subject: Re: [ovs-dev] [v4 03/12] dpif-netdev: Add study function to select
> the best
> mfex
Hi Eelco,
Thanks for you comment. My reply is inline.
Cian
> -Original Message-
> From: Eelco Chaudron
> Sent: Tuesday 29 June 2021 09:25
> To: Ferriter, Cian ; Van Haaren, Harry
>
> Cc: ovs-dev@openvswitch.org; i.maxim...@ovn.org; Flavio Leitner
>
> Subject: Re: [ovs-dev] [v13
Introduce build_lswitch_flows_for_lb routine in order to visit first
each load_balancer and then related datapath (logical switches) during
lb flow installation.
This patch allows to reduce memory footprint and cpu utilization in
ovn-northd.
Signed-off-by: Lorenzo Bianconi
---
Move stateful lb rules for logical switches in
build_lswitch_flows_for_lb routine in order to reduce cpu utilization
Signed-off-by: Lorenzo Bianconi
---
northd/ovn-northd.c | 42 ++
1 file changed, 18 insertions(+), 24 deletions(-)
diff --git
As it has been already done for logical routers, add logical switches
datapath references in ovn_northd_lb data structure.
This is a preliminary patch to invert the logic used during the lb flow
creation in order to visit lb first and then related datapath.
Signed-off-by: Lorenzo Bianconi
---
Remove add_router_lb_flow routine and move leftover lb flow
installation code in build_lrouter_snat_flows_for_lb routine
Signed-off-by: Lorenzo Bianconi
---
northd/ovn-northd.c | 258
1 file changed, 119 insertions(+), 139 deletions(-)
diff --git
Remove if condition that is never executed in
build_lrouter_nat_defrag_and_lb routine
Signed-off-by: Lorenzo Bianconi
---
northd/ovn-northd.c | 7 ---
1 file changed, 7 deletions(-)
diff --git a/northd/ovn-northd.c b/northd/ovn-northd.c
index 2e69394b3..d1507d4b5 100644
---
Introduce build_lrouter_snat_flows_for_lb routine to configuring
lb_{skip,force}_snat flows for each configured load_balancer
Signed-off-by: Lorenzo Bianconi
---
northd/ovn-northd.c | 205
1 file changed, 152 insertions(+), 53 deletions(-)
diff
Introduce build_lrouter_flows_for_lb routine in order to visit first each
load_balancer and then related datapath during lb flow installation.
This patch allows to reduce memory footprint and cpu utilization in
ovn-northd.
Signed-off-by: Lorenzo Bianconi
---
northd/ovn-northd.c | 109
add logical routers datapath references in ovn_northd_lb data structure.
This is a preliminary patch to invert the logic used during the lb flow
creation in order to visit lb first and then related datapath.
Signed-off-by: Lorenzo Bianconi
---
lib/lb.c| 11 +++
lib/lb.h
Move snat_type out of vip loop in build_lrouter_lb_flows() since there
is not vip dependency
Signed-off-by: Lorenzo Bianconi
---
northd/ovn-northd.c | 14 +-
1 file changed, 5 insertions(+), 9 deletions(-)
diff --git a/northd/ovn-northd.c b/northd/ovn-northd.c
index
Rework lb flow logic in order to visit first each load_balancer and then
related datapath during lb flow installation.
This patch allows to reduce memory footprint and cpu utilization in
ovn-northd.
Testing environment:
ovn-nbctl lr-list |wc -l
308
ovn-nbctl ls-list |wc -l
615
ovn-nbctl lb-list
> -Original Message-
> From: Eelco Chaudron
> Sent: Wednesday, June 30, 2021 10:18 AM
> To: Van Haaren, Harry
> Cc: Amber, Kumar ; d...@openvswitch.org;
> i.maxim...@ovn.org
> Subject: Re: [ovs-dev] [v4 03/12] dpif-netdev: Add study function to select
> the best
> mfex function
>
>
>
On 29 Jun 2021, at 20:49, Amber, Kumar wrote:
> Hi Flavio,
>
> Comments Inline.
>
>> -Original Message-
>> From: Flavio Leitner
>> Sent: Tuesday, June 29, 2021 11:49 PM
>> To: Amber, Kumar
>> Cc: Eelco Chaudron ; d...@openvswitch.org;
>> i.maxim...@ovn.org
>> Subject: Re: [ovs-dev]
On 29 Jun 2021, at 18:32, Van Haaren, Harry wrote:
>> -Original Message-
>> From: dev On Behalf Of Eelco Chaudron
>> Sent: Tuesday, June 29, 2021 1:38 PM
>> To: Amber, Kumar
>> Cc: d...@openvswitch.org; i.maxim...@ovn.org
>> Subject: Re: [ovs-dev] [v4 03/12] dpif-netdev: Add study
On 5/20/21 3:35 PM, Gaetan Rivet wrote:
> Joining pthreads makes the caller quiescent. It should register as such,
> as joined threads may wait on an RCU callback executing before quitting,
> deadlocking the caller.
Hi, Gaetan.
This patch doesn't look right to me. The problem is that users of
On 29 Jun 2021, at 18:15, Amber, Kumar wrote:
> Hi Eelco ,
>
> Sorry the formatting seems broken on this email thread.
> Replies are inlined .
Thanks, looking forward to the v5 (hopefully once I finished this version of
the series. I’m currently at patch 10 ;)
> From: Eelco Chaudron
> Sent:
On 29 Jun 2021, at 17:23, Van Haaren, Harry wrote:
>> -Original Message-
>> From: Eelco Chaudron
>> Sent: Tuesday, June 29, 2021 2:56 PM
>> To: Amber, Kumar
>> Cc: Van Haaren, Harry ; d...@openvswitch.org;
>> i.maxim...@ovn.org; Flavio Leitner
>> Subject: Re: [ovs-dev] [v4 01/12]
On 6/29/21 8:51 PM, Ben Pfaff wrote:
> This makes it easier to test northd behavior with particular database
> contents, like the ones that Dumitru posted to the mailing list:
> https://mail.openvswitch.org/pipermail/ovs-dev/2021-June/384519.html
>
> You just do something like this:
> make
On 6/29/21 6:04 PM, Paolo Valerio wrote:
> Dumitru Ceara writes:
>
>> On 6/25/21 2:01 PM, Paolo Valerio wrote:
>>> Dumitru Ceara writes:
>>>
On 6/21/21 12:06 PM, Paolo Valerio wrote:
> when a packet gets dnatted and then recirculated, it could be possible
> that it matches another
On 29 Jun 2021, at 22:43, Vasu Dasari wrote:
> Currently there is an option to add/flush/show ARP/ND neighbor. This covers L3
> side. For L2 side, there is only fdb show command. This patch gives an option
> to add/del an fdb entry via ovs-appctl.
>
> CLI command looks like:
>
> To add:
>
89 matches
Mail list logo