[ovs-dev] [PATCH v3 0/7] Improve linux QoS for exotic and fast links

There are currently two issues that limit our ability to configure QoS on certain cards in linux: 1) Firstly, the maximum link speed (which is used as maximum rate in some tc classes), is obtained from netdev feature bits. This is quite problematic because netdev feature bits do not keep up with t

[ovs-dev] [PATCH v3 1/7] netdev: add netdev_get_speed() to nedev API

Currently, the netdev's speed is being calculated by taking the link's feature bits (using netdev_get_features()) and transforming them into bps. This mechanism can be both inaccurate and difficult to maintain, mainly because we currently use the feature bits supported by OpenFlow which would have

[ovs-dev] [PATCH v3 3/7] netdev-linux: use 64bit rtab and burst calculations

tc uses these "rtab" tables to estimate the time (ticks) that it takes to send a packet of different sizes. In preparation for the introduction of 64-bit rates, add an argument to tc_put_rtab() to allow an external 64-bit rate. Also use 64bits for other burst buffer calculation functions. Signed-

[ovs-dev] [PATCH v3 4/7] netdev-linux: use 64-bit rates in htb tc classes

Currently, htb rates are capped at ~34Gbps because they are internally expressed as 32-bit fields. Move min and max rates to 64-bit fields and use TCA_HTB_RATE64 and TCA_HTB_CEIL64 to configure HTC classes to break this barrier. In order to test this, create a dummy tuntap device and set it's spe

[ovs-dev] [PATCH v3 2/7] netdev-linux: use speed as max rate in tc classes

Instead of relying on feature bits, use the speed value directly as maximum rate for htb and hfsc classes. There is still a limitation with the maximum rate that we can express with a 32-bit number in bytes/s (~ 34.3Gbps), but using the actual link speed instead of the feature bits, we can at leas

[ovs-dev] [PATCH v3 7/7] netdev-linux: support 64-bit rates in tc policing

Use TCA_POLICE_RATE64 if the rate cannot be expressed using 32bits. This breaks the 32Gbps barrier. The new barrier is ~4Tbps caused by netdev's API expressing kbps rates using 32-bit integers. Acked-by: Eelco Chaudron Reported-at: https://bugzilla.redhat.com/show_bug.cgi?id=2137643 Signed-off-b

[ovs-dev] [PATCH v3 6/7] netdev-linux: refactor nl_msg_put_act_police

In preparation for supporting 64-bit rates in tc policies, move the allocation and initialization of struct tc_police object inside nl_msg_put_act_police(). That way, the function is now called with the actual rates. Acked-by: Eelco Chaudron Reviewed-by: Simon Horman Signed-off-by: Adrian Moreno

[ovs-dev] [PATCH v3 5/7] netdev-linux: remove tc_matchall_fill_police

It is equivalent to tc_policer_init() so remove the duplicated function. Reviewed-by: Simon Horman Acked-by: Eelco Chaudron Signed-off-by: Adrian Moreno --- lib/netdev-linux.c | 27 +++ 1 file changed, 3 insertions(+), 24 deletions(-) diff --git a/lib/netdev-linux.c b/

Re: [ovs-dev] [PATCH net-next 2/2] net: openvswitch: add drop action

On 7/11/23 22:46, Aaron Conole wrote: Eric Garver writes: On Mon, Jul 10, 2023 at 06:51:19PM +0200, Ilya Maximets wrote: On 7/8/23 00:06, Jakub Kicinski wrote: On Fri, 7 Jul 2023 18:04:36 +0200 Ilya Maximets wrote: That already exists, right? Johannes added it in the last release for WiF

[ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

Hi all, During the ovn-heater community meeting organized by Frode yesterday (thanks again for that!) we agreed to follow up on some points. Two of these are related to gathering more information that allow us to build realistic test scenarios (1) and define targets for them (2). On that note we

Re: [ovs-dev] [ovn] ovn-controller high memory consumption with sbrec_server_has_##table##_table - enabled codepath

On 7/12/23 00:01, Ilya Maximets wrote: > On 7/11/23 19:01, Dumitru Ceara wrote: >> On 7/11/23 18:33, Vladislav Odintsov wrote: >>> Hi Dumitru, >>> >>> The system on which I reproduced this issue is running 22.09.x version. >>> I’ve tried to upgrade ovn-controller to main branch + your patch. Pleas

[ovs-dev] [PATCH v4] conntrack: Extract l4 information for SCTP.

since a27d70a89 ("conntrack: add generic IP protocol support") all the unrecognized IP protocols get handled using ct_proto_other ops and are managed as L3 using 3 tuples. This patch stores L4 information for SCTP in the conn_key so that multiple conn instances, instead of one with ports zeroed, w

Re: [ovs-dev] Scale testing OVN with ovn-heater for OpenStack use cases

On Mon, Jul 10, 2023 at 9:42 AM Frode Nordahl wrote: > > Have now sent out an invite to everyone participating in this thread. > > For the benefit of anyone else wanting to attend I'm also sharing the > video link and other resources here: > Date/Time: Tuesday July 11th 13:30 UTC > Video link: htt

Re: [ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

On Wed, Jul 12, 2023 at 10:51 AM Dumitru Ceara wrote: > > Hi all, > > During the ovn-heater community meeting organized by Frode yesterday > (thanks again for that!) we agreed to follow up on some points. Two of > these are related to gathering more information that allow us to build > realistic

Re: [ovs-dev] [PATCH v2 3/3] checkpatch: print subject field if misspelled or missing

On 11 Jul 2023, at 22:10, Chandan Somani wrote: > On Tue, Jul 11, 2023 at 12:38 AM Eelco Chaudron wrote: > >> >> >> On 7 Jul 2023, at 22:07, Chandan Somani wrote: >> >>> This narrows down spelling errors that are in the commit >>> subject. In v2, it also provides a subject if the subject >>> li

Re: [ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

On 7/12/23 12:04, Frode Nordahl wrote: > On Wed, Jul 12, 2023 at 10:51 AM Dumitru Ceara wrote: >> >> Hi all, >> >> During the ovn-heater community meeting organized by Frode yesterday >> (thanks again for that!) we agreed to follow up on some points. Two of >> these are related to gathering more

Re: [ovs-dev] [PATCH ovn] Remove lport from related_lports if there is no binding

On 6/6/23 14:10, Xavier Simonart wrote: > Hi Priyankar, Mark > > Thanks for the patch. I agree with Mark - the description is really great ! > Based on your description, I tried creating a unit-test reproducing the > issue, and checking that your patch fixes it. > I came up with [0]. It reproduces

Re: [ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

Hi everyone, On Wed, Jul 12, 2023 at 12:29:58PM +0200, Dumitru Ceara wrote: > On 7/12/23 12:04, Frode Nordahl wrote: > > On Wed, Jul 12, 2023 at 10:51 AM Dumitru Ceara wrote: > >> > >> Hi all, > >> > >> During the ovn-heater community meeting organized by Frode yesterday > >> (thanks again for th

Re: [ovs-dev] [PATCH v2] relay: allow setting probe interval

On 7/10/23 08:19, Felix Huettner via dev wrote: > previously it was not possible to set the probe interval for the > connection from a relay to the backing ovsdb-server. With this change it > is now possible using the > `ovsdb-server/set-active-ovsdb-server-probe-interval` command. > > The command

Re: [ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

On 7/12/23 13:00, Felix Huettner wrote: > Hi everyone, > > On Wed, Jul 12, 2023 at 12:29:58PM +0200, Dumitru Ceara wrote: >> On 7/12/23 12:04, Frode Nordahl wrote: >>> On Wed, Jul 12, 2023 at 10:51 AM Dumitru Ceara wrote: Hi all, During the ovn-heater community meeting organiz

Re: [ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

On Wed, Jul 12, 2023 at 01:57:18PM +0200, Dumitru Ceara wrote: > On 7/12/23 13:00, Felix Huettner wrote: > > Hi everyone, > > > > On Wed, Jul 12, 2023 at 12:29:58PM +0200, Dumitru Ceara wrote: > >> On 7/12/23 12:04, Frode Nordahl wrote: > >>> On Wed, Jul 12, 2023 at 10:51 AM Dumitru Ceara wrote: >

Re: [ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

On Wed, Jul 12, 2023 at 2:31 PM Felix Huettner wrote: > > On Wed, Jul 12, 2023 at 01:57:18PM +0200, Dumitru Ceara wrote: > > On 7/12/23 13:00, Felix Huettner wrote: > > > Hi everyone, > > > > > > On Wed, Jul 12, 2023 at 12:29:58PM +0200, Dumitru Ceara wrote: > > >> On 7/12/23 12:04, Frode Nordahl

[ovs-dev] [PATCH v7] ofproto-dpif-xlate: Reduce stack usage in recursive xlate functions

Several xlate actions used in recursive translation currently store a large amount of information on the stack. This can result in handler threads quickly running out of stack space despite before xlate_resubmit_resource_check() is able to terminate translation. This patch reduces stack usage by ov

Re: [ovs-dev] [PATCH v7] ofproto-dpif-xlate: Reduce stack usage in recursive xlate functions

On 12 Jul 2023, at 15:37, Mike Pattrick wrote: > Several xlate actions used in recursive translation currently store a > large amount of information on the stack. This can result in handler > threads quickly running out of stack space despite before > xlate_resubmit_resource_check() is able to

Re: [ovs-dev] [PATCH v3 7/7] netdev-linux: support 64-bit rates in tc policing

On 12 Jul 2023, at 9:02, Adrian Moreno wrote: > Use TCA_POLICE_RATE64 if the rate cannot be expressed using 32bits. > > This breaks the 32Gbps barrier. The new barrier is ~4Tbps caused by > netdev's API expressing kbps rates using 32-bit integers. > > Acked-by: Eelco Chaudron > Reported-at: htt

Re: [ovs-dev] [PATCH v3 3/7] netdev-linux: use 64bit rtab and burst calculations

On 12 Jul 2023, at 9:02, Adrian Moreno wrote: > tc uses these "rtab" tables to estimate the time (ticks) that it takes > to send a packet of different sizes. In preparation for the introduction > of 64-bit rates, add an argument to tc_put_rtab() to allow an external > 64-bit rate. > > Also use

Re: [ovs-dev] [PATCH v3 1/7] netdev: add netdev_get_speed() to nedev API

On 12 Jul 2023, at 9:02, Adrian Moreno wrote: > Currently, the netdev's speed is being calculated by taking the link's > feature bits (using netdev_get_features()) and transforming them into > bps. > > This mechanism can be both inaccurate and difficult to maintain, mainly > because we currently

Re: [ovs-dev] [PATCH v3 2/7] netdev-linux: use speed as max rate in tc classes

On 12 Jul 2023, at 9:02, Adrian Moreno wrote: > Instead of relying on feature bits, use the speed value directly as > maximum rate for htb and hfsc classes. > > There is still a limitation with the maximum rate that we can express > with a 32-bit number in bytes/s (~ 34.3Gbps), but using the ac

Re: [ovs-dev] [PATCH v3 4/7] netdev-linux: use 64-bit rates in htb tc classes

On 12 Jul 2023, at 9:02, Adrian Moreno wrote: > Currently, htb rates are capped at ~34Gbps because they are internally > expressed as 32-bit fields. > > Move min and max rates to 64-bit fields and use TCA_HTB_RATE64 and > TCA_HTB_CEIL64 to configure HTC classes to break this barrier. > > In ord

Re: [ovs-dev] [PATCH ovn v2 1/8] northd I-P: Sync SB load balancers in a separate engine node.

On Fri, Jul 7, 2023 at 1:53 PM wrote: > > From: Numan Siddique > > Similar to the commit [1], a new sub-engine node "sync_to_sb_lb" > is added with-in the "sync_to_sb" to sync the SB load balancers. > Its main input nodes are "northd" (to access the "lbs" hmap built > by this node) and "sb_load_b

Re: [ovs-dev] [PATCH ovn v2 2/8] northd: Add a new engine node - northd_lb_data.

On Sat, Jul 8, 2023 at 3:57 AM Mark Michelson wrote: > > Hi Numan, > > I have one small nit below. +1 Acked-by: Han Zhou > > On 7/7/23 01:53, num...@ovn.org wrote: > > From: Numan Siddique > > > > This patch separates out the 'lbs' and 'lb_groups' from the 'northd' engine > > node data into a

Re: [ovs-dev] [PATCH ovn v2 3/8] northd: Add initial I-P for load balancer and load balancer groups

On Sat, Jul 8, 2023 at 3:57 AM Mark Michelson wrote: > > Hi Numan, > > I have one small nit below. > > On 7/7/23 01:53, num...@ovn.org wrote: > > From: Numan Siddique > > > > Any changes to load balancers and load balancer groups > > are handled incrementally in the newly added 'northd_lb_data' >

Re: [ovs-dev] [PATCH ovn v2 0/8] northd: I-P for load balancer and lb groups

On Tue, Jul 11, 2023 at 10:41 PM Numan Siddique wrote: > > On Sat, Jul 8, 2023 at 1:27 AM Mark Michelson wrote: > > > > Hi Numan, > > > > I gave the series a look. I've looked at the code but haven't yet run > > any tests with it. The main reason for this is that the series does not > > apply cle

Re: [ovs-dev] [PATCH] ovsdb: raft: Support pre-vote mechanism to deal with disruptive server.

On 7/7/23 07:14, Han Zhou wrote: > > > On Mon, Jul 3, 2023 at 9:46 PM Ilya Maximets > wrote: >> >> On 7/1/23 04:43, Han Zhou wrote: >> > >> > >> > On Fri, Jun 30, 2023 at 4:30 PM Ilya Maximets > > > >

Re: [ovs-dev] [PATCH ovn v2] tests: fixed userspace-system tests not properly cleaned up

On 7/7/23 10:46, Ales Musil wrote: > On Wed, Jul 5, 2023 at 5:57 PM Xavier Simonart wrote: > >> Usually, when cleaning up a system test we execute (for the ovn-vswitchd >> process): >> - OVS_TRAFFIC_VSWITCHD_STOP (i.e. ovs-appctl -t $1 exit --cleanup, wait up >> to 30 seconds >> and report fail

Re: [ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

On 7/12/23 14:42, Frode Nordahl wrote: > On Wed, Jul 12, 2023 at 2:31 PM Felix Huettner > wrote: >> >> On Wed, Jul 12, 2023 at 01:57:18PM +0200, Dumitru Ceara wrote: >>> On 7/12/23 13:00, Felix Huettner wrote: Hi everyone, On Wed, Jul 12, 2023 at 12:29:58PM +0200, Dumitru Ceara wrot

Re: [ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

Hi Dumitru, first of all, I’l like to mention that’s a great idea! I’m wondering wether ovn interconnection infrastructure and ovn vtep could also be a part of this testing? On 12 Jul 2023, at 22:38, Dumitru Ceara wrote: On 7/12/23 14:42, Frode Nordahl wrote: On Wed, Jul 12, 2023 at 2:31 PM Fe

Re: [ovs-dev] [PATCH v4 1/2] debian: Build package with AF_XDP.

On 6/27/23 12:11, Frode Nordahl wrote: > Build the upstream Open vSwitch deb package with AF_XDP datapath s/datapath/ports/ > enabled when required dependencies are available. > > Signed-off-by: Frode Nordahl > --- > Documentation/intro/install/debian.rst | 13 > debian/automake.mk

Re: [ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

On 7/12/23 21:46, Odintsov Vladislav wrote: > Hi Dumitru, > Hi Vladislav, > first of all, I’l like to mention that’s a great idea! > I’m wondering wether ovn interconnection infrastructure and ovn vtep could > also be a part of this testing? > There's actually already work in progress to add

Re: [ovs-dev] [PATCH ovn] binding: fixed port claims as additional_chassis

On 7/7/23 14:11, Ales Musil wrote: > On Tue, Jun 27, 2023 at 11:37 AM Xavier Simonart > wrote: > >> When sb is read-only, the port claim is delayed until sb is rw. >> However, before this patch, this resulted in the chassis always >> claiming the port as main (while it was maybe an additional cha

Re: [ovs-dev] [RFC ovn-heater] OVN scale testing with OpenStack workloads - questionnaire

On 12 Jul 2023, at 23:09, Dumitru Ceara wrote: On 7/12/23 21:46, Odintsov Vladislav wrote: Hi Dumitru, Hi Vladislav, first of all, I’l like to mention that’s a great idea! I’m wondering wether ovn interconnection infrastructure and ovn vtep could also be a part of this testing? There's ac

Re: [ovs-dev] [PATCH ovn] northd: Lookup route policy before ip routing

Hi wangchuanlei, Thanks for the patch! On 7/10/23 10:46, wangchuanlei wrote: > If there is no route in table ip_routing, the route policy > item in table policy is useless. I'm sorry but I disagree with this. We can always add a default route that matches all traffic that's not matched by a

Re: [ovs-dev] [PATCH v4 2/2] debian: Run system testsuites as autopkgtest.

On 6/27/23 12:11, Frode Nordahl wrote: > The autopkgtests [0][1] are relevant in an upstream context > because an Open vSwitch contributor may want to have a quick > way of running the upstream system testsuites on recent > Debian/Ubuntu releases in an automated and contained manner. > > During th

Re: [ovs-dev] [PATCH ovn v2 2/8] northd: Add a new engine node - northd_lb_data.

On Wed, Jul 12, 2023 at 11:54 PM Han Zhou wrote: > > > > On Sat, Jul 8, 2023 at 3:57 AM Mark Michelson wrote: > > > > Hi Numan, > > > > I have one small nit below. > > +1 > Acked-by: Han Zhou > Sorry that I forgot one small comment regarding the naming of the new node (and the related files). I

Re: [ovs-dev] [PATCH ovn v2 4/8] northd: Refactor the 'northd' node code which handles logical switch changes.

On Fri, Jul 7, 2023 at 1:54 PM wrote: > > From: Numan Siddique > > This will help in handling other column changes of a logical switch. > > Signed-off-by: Numan Siddique > --- > northd/northd.c | 260 > 1 file changed, 151 insertions(+), 109 dele