Re: [ovs-dev] [ovs-discuss] OVS 2.12/2.13 compilation on Ubuntu Bionic

Hello, Thank you for your responses! Is there any reason not to use the in-tree openvswitch kernel module > provided in the Ubuntu kernels? Ubuntu stopped shipping DKMS modules as > part of OVS quite a long time ago as the openvswitch module in the kernel > is well maintained and generally up-to

[ovs-dev] please help me to confirm whether there is memleak in ovs-vswitchd

Hello, It seems there is memory leak in ovs-vswitchd. everything is going fine while I enabled dpdk in ovs-vswitchd and created netdev bridge, but if I add dpdk ports to netdev bridge, ovs-vswitchd starts to consume memory until the system runs out of memory, I’m using ovs-2.13.0

[ovs-dev] [PATCH v1 1/2] netdev-dpdk: Add ability to set MAC address.

From: Ilya Maximets From: Ilya Maximets It is possible to set MAC address for DPDK ports by calling rte_eth_dev_default_mac_addr_set(). For some reason OVS didn't use this functionality avoiding real MAC address configuration. With this change following command will result in real MAC address

[ovs-dev] [PATCH v1 0/2] netdev-dpdk: support changing VF MAC

Hello Ilya, Following your suggestion, here is a small patch adding the ability to configure the MAC address of DPDK VF representors. As said off-ML, I've used the options column instead of other_configs, as it allows avoid having the DPDK representor concept bleed out of netdev-dpdk. There is

[ovs-dev] [PATCH v1 2/2] netdev-dpdk: Add option to configure VF MAC address.

In some cloud topologies, using DPDK VF representors in guest requires configuring a VF before it is assigned to the guest. A first basic option for such configuration is setting the VF MAC address. Add a key 'dpdk-vf-mac' to the 'options' column of the Interface table. This option can be used as

[ovs-dev] お支払い方法の情報を更新

お支払い方法の情報を更新してください。Update default card for your membership. マイストア? タイムセール? ギフト券 Amazonプライムをご利用頂きありがとうございます。お客様のAmazonプライム会員資格は、2020/06/09に更新を迎えます。お調べしたところ、会費のお支払いに使用できる有効なクレジットカードがアカウントに登録されていません。クレジットカード情報の更新、新しいクレジットカードの追加については以下の手順をご確認ください。 1. アカウントサービスからAmazonプライム会員情報を管理するにアクセスします。

Re: [ovs-dev] [PATCH ovn v5] ovn-northd: Minimize number of ARP/NS responder flows for DNAT.

On 7/7/20 12:55 AM, Han Zhou wrote: > > > On Mon, Jul 6, 2020 at 4:33 AM Dumitru Ceara > wrote: >> >> Most ARP/NS responder flows can be configured per datapath instead of per >> router port. >> >> The only exception is with distributed gateway router ports which need >

Re: [ovs-dev] [PATCH ovn 2/3] Introduce icmp6.frag_mtu action

On Fri, Jul 3, 2020 at 7:55 PM Lorenzo Bianconi wrote: > Similar to what have been already done for IPv4, introduce > icmp6.frag_mtu action in order to set correct mtu in ICMPv6 "packet too > big" error message > > Signed-off-by: Lorenzo Bianconi > Hi Lorenzo, This patch has a compilation warn

Re: [ovs-dev] [PATCH v1 2/2] netdev-dpdk: Add option to configure VF MAC address.

Bleep bloop. Greetings Gaetan Rivet, 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 92 characters long (recommended limit is 79) #131 FILE: lib/netdev-dpdk.c:2016

Re: [ovs-dev] [PATCH v3 3/4] netdev-offload: Add xdp flow api provider

On 2020/06/30 0:30, Toshiaki Makita wrote: ... int netdev_afxdp_init(void) { libbpf_set_print(libbpf_print); -return 0; +return netdev_register_flow_api_provider(&netdev_offload_xdp); This causes duplicate flow api provider error because afxdp and afxdp-nonpmd are using the s

Re: [ovs-dev] [PATCH v3 0/4] XDP offload using flow API provider

On 2020/06/30 0:30, Toshiaki Makita wrote: This patch adds an XDP-based flow cache using the OVS netdev-offload flow API provider. When an OVS device with XDP offload enabled, packets first are processed in the XDP flow cache (with parse, and table lookup implemented in eBPF) and if hits, the ac

Re: [ovs-dev] [PATCH v2] ofproto: delete buckets when lb_output is false

On 7/7/20 8:45 AM, Adrian Moreno wrote: > When lb-output-action is toggled back to "false" buckets are not being > deleted. Delete them as they will no longer be used. > > Add unit test to verify buckets are correctly deleted. > > Fixes: 9df65060cf4c ("userspace: Avoid dp_hash recirculation for b

[ovs-dev] [PATCH v2 0/2] netdev-dpdk: support changing VF MAC

v2: fix 0-day bot issues in 2/2. Hello Ilya, Following your suggestion, here is a small patch adding the ability to configure the MAC address of DPDK VF representors. As said off-ML, I've used the options column instead of other_configs, as it allows avoid having the DPDK representor concept b

[ovs-dev] [PATCH v2 1/2] netdev-dpdk: Add ability to set MAC address.

From: Ilya Maximets It is possible to set MAC address for DPDK ports by calling rte_eth_dev_default_mac_addr_set(). For some reason OVS didn't use this functionality avoiding real MAC address configuration. With this change following command will result in real MAC address update on HW NIC:

[ovs-dev] [PATCH v2 2/2] netdev-dpdk: Add option to configure VF MAC address.

In some cloud topologies, using DPDK VF representors in guest requires configuring a VF before it is assigned to the guest. A first basic option for such configuration is setting the VF MAC address. Add a key 'dpdk-vf-mac' to the 'options' column of the Interface table. This option can be used as

Re: [ovs-dev] [PATCH ovn 2/3] Introduce icmp6.frag_mtu action

On Tue, Jul 7, 2020 at 2:33 PM Numan Siddique wrote: > > > On Fri, Jul 3, 2020 at 7:55 PM Lorenzo Bianconi < > lorenzo.bianc...@redhat.com> wrote: > >> Similar to what have been already done for IPv4, introduce >> icmp6.frag_mtu action in order to set correct mtu in ICMPv6 "packet too >> big" err

Re: [ovs-dev] [PATCH v4 3/5] dpif-netdev: Skip encap action during datapath execution

On 7/6/2020 4:20 PM, Sriharsha Basavapatna wrote: On Sun, Jul 5, 2020 at 7:33 PM Eli Britstein wrote: On 6/29/2020 12:50 PM, Sriharsha Basavapatna wrote: In this patch we check if action processing (apart from OUTPUT action), should be skipped for a given dp_netdev_flow. Specifically, we ch

Re: [ovs-dev] [PATCH 1/2] ovsdb/TODO.rst: Remove completed items.

On 6/22/20 7:55 AM, Han Zhou wrote: > - snapshot unit test has been added for "change-election-timer" related > patches. > > - 100% CPU problem was addressed by: > 2cd62f75c1 ("ovsdb raft: Precheck prereq before proposing commit.") > > Signed-off-by: Han Zhou > --- Thanks! I applied both p

Re: [ovs-dev] [PATCH v3 ovn] Throttle the OVS-OVN Global config logs

On Sun, Jul 5, 2020 at 11:08 AM Ankur Sharma wrote: > From: Ankur Sharma > > ISSUE: > We observed that if ovn-controller is running, while ovn-encap-ip/ > ovn-encap-type is not set, then following error gets logged > continously: > "chassis|INFO|Need to specify an encap type and ip" > > Above lo

Re: [ovs-dev] [PATCH v8] ovsdb-idl: Force IDL retry when missing updates encountered.

On 7/2/20 4:20 PM, Dumitru Ceara wrote: > Adds a generic recovery mechanism which triggers an IDL retry with fast > resync disabled in case the IDL has detected that it ended up in an > inconsistent state due to other bugs in the ovsdb-server/ovsdb-idl > implementation. > > Additionally, this comm

Re: [ovs-dev] [ovs-discuss] OVS 2.12/2.13 compilation on Ubuntu Bionic

On Tue, Jul 7, 2020 at 8:39 AM Maciej Jozefczyk wrote: > Hello, > > Thank you for your responses! > > Is there any reason not to use the in-tree openvswitch kernel module >> provided in the Ubuntu kernels? Ubuntu stopped shipping DKMS modules as >> part of OVS quite a long time ago as the openvs

[ovs-dev] [PATCH V5 01/12] netdev-offload-dpdk: Log testpmd format for flow create/destroy

To enhance debugability with DPDK, format the logs in a testpmd format commands. Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai --- lib/netdev-offload-dpdk.c | 381 -- 1 file changed, 168 insertions(+), 213 deletions(-) diff --git a/lib/net

[ovs-dev] [PATCH V5 03/12] dpif-netdev: Don't use zero flow mark

Zero flow mark is used to indicate the HW to remove the mark. A packet marked with zero mark is received in SW without a mark at all, so it cannot be used as a valid mark. Change the pool range to fix it. Fixes: 241bad15d99a ("dpif-netdev: associate flow with a mark id") Signed-off-by: Eli Britste

[ovs-dev] [PATCH V5 07/12] netdev-offload-dpdk: Add IPv6 pattern matching

Add support for IPv6 pattern matching for offloading flows. Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai Acked-by: Sriharsha Basavapatna --- Documentation/howto/dpdk.rst | 2 +- NEWS | 1 + lib/netdev-offload-dpdk.c| 75

[ovs-dev] [PATCH V5 04/12] netdev-offload-dpdk: Fix Ethernet matching for type only

For OVS rule of the form "eth type is 0x1234 / end", rule is offloaded in the form of "eth / end", which is incorrect. Fix it. Fixes: e8a2b5bf92bb ("netdev-dpdk: implement flow offload with rte flow") Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai Acked-by: Sriharsha Basavapatna ---

[ovs-dev] [PATCH V5 02/12] dpif-netdev: Add mega ufid in flow add/del log

As offload is done using the mega ufid of a flow, for better debugability, add it in the log message. Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai --- lib/dpif-netdev.c | 13 + tests/dpif-netdev.at| 2 ++ tests/ofproto-macros.at | 3 ++- 3 files changed, 13 i

[ovs-dev] [PATCH V5 05/12] netdev-offload-dpdk: Support partial TCP/UDP port matching

The cited commit failed partial matching of TCP/UDP port matching, preventing such offload of supporting HWs. Remove this failure. Fixes: e8a2b5bf92bb ("netdev-dpdk: implement flow offload with rte flow") Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai Acked-by: Sriharsha Basavapatna -

[ovs-dev] [PATCH V5 00/12] netdev datapath offload: Support IPv6 and VXLAN encap

This patch set includes enhanced logging to increase debugability, bug fixes and additional offloads - IPv6 and VXLAN encap. Patches #1-#2: Enhance log prints for debugability. Patches #3-#4: Bug fixes for partial offloads and Ethernet matching. Patches #5-#12: Add support for offloads of IPv6 p

[ovs-dev] [PATCH V5 12/12] netdev-offload-dpdk: Support tnl/push using vxlan encap attribute

For DPDK, there is the RAW_ENCAP attribute which gets raw buffer of the encapsulation header. For specific protocol, such as vxlan, there is a more specific attribute, VXLAN_ENCAP, which gets the parsed fields of the outer header. In case tunnel type is vxlan, parse the header and use the specific

[ovs-dev] [PATCH V5 09/12] netdev: Allow storing dpif type into netdev structure.

From: Ilya Maximets Storing of the dpif type of the owning datapath interface will allow us to easily distinguish, for example, userspace tunneling ports from the system ones. This is required in terms of HW offloading to avoid offloading of userspace flows to kernel interfaces that doesn't belo

[ovs-dev] [PATCH V5 06/12] netdev-offload-dpdk: Remove pre-validate of patterns function

The function of adding patterns by requested matches checks that it consumed all the required matches, and err if not. For functional purpose there is no need for pre-validation. For performance such validation may decrease the time spent for failing flows, but at the expense of increasing the time

[ovs-dev] [PATCH V5 08/12] netdev-offload-dpdk: Support offload of set IPv6 actions

Add support for set IPv6 actions. Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai Acked-by: Sriharsha Basavapatna --- Documentation/howto/dpdk.rst | 1 + NEWS | 2 ++ lib/netdev-offload-dpdk.c| 32 3 files changed, 35 inse

[ovs-dev] [PATCH V5 10/12] netdev-offload: Use dpif type instead of class.

From: Ilya Maximets There is no real difference between the 'class' and 'type' in the context of common lookup operations inside netdev-offload module because it only checks the value of pointers without using the value itself. However, 'type' has some meaning and can be used by offload provides

[ovs-dev] [PATCH V5 11/12] netdev-offload-dpdk: Support offload of clone tnl_push/output actions

Tunnel encapsulation is done by tnl_push and output actions nested in a clone action. Support offloading of such flows with RTE_FLOW_ACTION_TYPE_RAW_ENCAP attribute. Signed-off-by: Eli Britstein Reviewed-by: Oz Shlomo Acked-by: Sriharsha Basavapatna --- Documentation/howto/dpdk.rst | 1 + NEW

[ovs-dev] [PATCH ovn] ovn-northd: Fix the missing lflow issue in LS_OUT_PRE_LB.

From: Numan Siddique When load balancer(s) configured with VIPs are associated to a logical switch, then ovn-northd adds the below logical flow so that the packets in the egress switch pipeline enter the conntrack. table=0 (ls_out_pre_lb ), priority=100 , match=(ip), action=(reg0[[0]] = 1

Re: [ovs-dev] [PATCH v5] system-dpdk: add negotiation check for userspace-tso

Hi Gowrishankar, Thanks for following up with this patch. However, I think we need to reconsider the tests. It doesn't seem useful to have negotiation tests that will always fail because of a code that is marked as deprecated (vhost-user). I see few options here: 1) Fix vhost-user to support

[ovs-dev] [PATCH V2 RFC dpdk-latest] build: Add support for DPDK meson build.

Make based build is deprecated in DPDK. Meson based build to be used for future DPDK releases. This updates travis, configure script and documentation for using DPDK Meson with OVS. Tested-at: https://travis-ci.org/github/Sunil-Pai-G/ovs/builds/705746816 Signed-off-by: Sunil Pai G --- v1->v2:

Re: [ovs-dev] [PATCH ovn] ovn-northd: Fix the missing lflow issue in LS_OUT_PRE_LB.

On 7/7/20 3:16 PM, num...@ovn.org wrote: > From: Numan Siddique > > When load balancer(s) configured with VIPs are associated to a logical switch, > then ovn-northd adds the below logical flow so that the packets in the egress > switch pipeline enter the conntrack. > > table=0 (ls_out_pre_lb

Re: [ovs-dev] [PATCH ovn 2/3] Introduce icmp6.frag_mtu action

> On Fri, Jul 3, 2020 at 7:55 PM Lorenzo Bianconi > wrote: > [...] > > > > Looks like this was used for debugging and you forgot to clean it up ? > > And the compilation warning is because of this. Ops..sorry about this, I will fix it in v2 > > > > > +/* compute checksum and set

Re: [ovs-dev] [PATCH v1 0/6] Memory access optimization for flow scalability of userspace datapath.

> -Original Message- > From: dev On Behalf Of Yanqin Wei > Sent: Tuesday, June 2, 2020 8:10 AM > To: d...@openvswitch.org > Cc: ruifeng.w...@arm.com; lijian.zh...@arm.com; i.maxim...@ovn.org; > n...@arm.com > Subject: [ovs-dev] [PATCH v1 0/6] Memory access optimization for flow > scalabil

Re: [ovs-dev] [PATCH v8] ovsdb-idl: Force IDL retry when missing updates encountered.

On 7/7/20 1:01 PM, Ilya Maximets wrote: > On 7/2/20 4:20 PM, Dumitru Ceara wrote: >> Adds a generic recovery mechanism which triggers an IDL retry with fast >> resync disabled in case the IDL has detected that it ended up in an >> inconsistent state due to other bugs in the ovsdb-server/ovsdb-idl >

Re: [ovs-dev] [PATCH ovn] ovn-northd: Fix the missing lflow issue in LS_OUT_PRE_LB.

On Tue, Jul 7, 2020 at 8:03 PM Dumitru Ceara wrote: > On 7/7/20 3:16 PM, num...@ovn.org wrote: > > From: Numan Siddique > > > > When load balancer(s) configured with VIPs are associated to a logical > switch, > > then ovn-northd adds the below logical flow so that the packets in the > egress > >

[ovs-dev] [PATCH v2 ovn 1/3] Introduce icmp6_error action

Intoduce icmp6_error action in order to generate an ICMPv6 packet in response to an error in original IPv6 packet Signed-off-by: Lorenzo Bianconi --- controller/pinctrl.c | 1 + include/ovn/actions.h | 9 - lib/actions.c | 22 ++ ovn-sb.xml| 8

[ovs-dev] [PATCH v2 ovn 2/3] Introduce icmp6.frag_mtu action

Similar to what have been already done for IPv4, introduce icmp6.frag_mtu action in order to set correct mtu in ICMPv6 "packet too big" error message Signed-off-by: Lorenzo Bianconi --- controller/pinctrl.c | 72 ++-- include/ovn/actions.h| 4 ++

[ovs-dev] [PATCH v2 ovn 0/3] enable IPv6 PMTU discovery

Changes since v1: - remove leftover debug comment - fix icmp6.frag_mtu documentation Lorenzo Bianconi (3): Introduce icmp6_error action Introduce icmp6.frag_mtu action northd: introduce icmp6_error logical flows in router pipeline controller/pinctrl.c | 73 +++--

[ovs-dev] [PATCH v2 ovn 3/3] northd: introduce icmp6_error logical flows in router pipeline

Introduce icmp6_error logical flows in router pipeline if gateway_mtu has been added to logical router port option column in order to perform IPv6 PMTU discovery Signed-off-by: Lorenzo Bianconi --- northd/ovn-northd.8.xml | 24 +--- northd/ovn-northd.c | 84 ++

Re: [ovs-dev] [PATCH V2] dpif-netdev: Do RCU synchronization at fixed interval in PMD main loop.

On 7/2/20 2:36 PM, Ilya Maximets wrote: > On 11/12/19 7:38 AM, Nitin Katiyar wrote: >> Hi Guys, >> Can you please review the patch? It has been in mailing list for long time. > > Hi. > > Really sorry for so long delay. I lost track of this change somehow during > my relocation and job changes. >

Re: [ovs-dev] [PATCH] python: Fix plural forms of OVSDB types.

On Tue, Jul 07, 2020 at 06:11:33AM +0530, Numan Siddique wrote: > On Tue, Jul 7, 2020, 2:38 AM Ben Pfaff wrote: > > > This still needs a review. > > > > Oops. Sorry about that. > > Acked-by: Numan Siddique Thanks! Applied to master. ___ dev mailing

[ovs-dev] [PATCH ovn 1/2] ovn-detrace: Add support for multiple remotes.

The --ovnnb/--ovnsb/--ovsdb arguments now accept a list of remotes for the case when the databases are clustered. Signed-off-by: Dumitru Ceara --- utilities/ovn-detrace.in | 21 +++-- 1 file changed, 15 insertions(+), 6 deletions(-) diff --git a/utilities/ovn-detrace.in b/util

[ovs-dev] [PATCH ovn 0/2] Improve ovn-detrace usability with clustered DBs.

Patch 1 of the series adds support for multiple remotes when connecting to OVSDBs. This is useful when the NB/SB DBs are clustered. Patch 2 of the series adds support for connecting to DB remotes via SSL. Signed-off-by: Dumitru Ceara Dumitru Ceara (2): ovn-detrace: Add support for multipl

[ovs-dev] [PATCH ovn 2/2] ovn-detrace: Support SSL remotes.

Additional command line arguments are added to ovn-detrace to allow connecting to NB/SB/OVS DBs via SSL. Signed-off-by: Dumitru Ceara --- utilities/ovn-detrace.1.in | 30 ++ utilities/ovn-detrace.in | 38 +++--- 2 files changed, 6

[ovs-dev] [PATCH v3 0/4] Use more sensitive terminology.

There's no value in risking turning people away from OVS due to insensitive terminology choices. We can do better. v1->v2: Rebase. v2->v3: Fix some new instances of "slave" that had crept in (thanks Ilya!). Drop a patch that I had accidentally appended in v2. Ben Pfaff (4): Eliminate use of

[ovs-dev] [PATCH v3 4/4] Eliminate "whitelist" and "blacklist" terms.

There is one remaining use under datapath. That change should happen upstream in Linux first according to our usual policy. Signed-off-by: Ben Pfaff --- Documentation/howto/ipsec.rst | 2 +- Documentation/howto/selinux.rst | 8 +-- Documentation/topics/datapath.r

[ovs-dev] [PATCH v3 3/4] conntrack: Rename "master" connection to "primary" connection.

Signed-off-by: Ben Pfaff --- include/openvswitch/meta-flow.h | 14 +++ lib/conntrack-private.h | 10 ++--- lib/conntrack.c | 74 - lib/ct-dpif.c | 6 +-- lib/ct-dpif.h | 2 +- lib/meta-flow.xml

[ovs-dev] [PATCH v3 2/4] Use primary/secondary, not master/slave, as names for OpenFlow roles.

Signed-off-by: Ben Pfaff --- Documentation/topics/design.rst | 110 +++--- NEWS | 2 + include/openflow/nicira-ext.h| 22 ++--- include/openflow/openflow-1.2.h | 6 +- include/openflow/openflow-1.4.h | 8 +- include/ope

Re: [ovs-dev] [PATCH v2 1/5] Eliminate use of term "slave" in bond, LACP, and bundle contexts.

On Tue, Jul 07, 2020 at 01:57:42AM +0200, Ilya Maximets wrote: > On 7/6/20 10:35 PM, Ben Pfaff wrote: > > Most of these changes should not change user-visible behavior. One > > place where they do is in "ovs-ofctl dump-flows", which will now output > > "subs:..." inside "bundle" actions instead of

Re: [ovs-dev] [PATCH v3 1/4] Eliminate use of term "slave" in bond, LACP, and bundle contexts.

Bleep bloop. Greetings Ben Pfaff, 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 91 characters long (recommended limit is 79) #1866 FILE: lib/lacp.c:1160: lacp_get

Re: [ovs-dev] [PATCH v3 2/4] Use primary/secondary, not master/slave, as names for OpenFlow roles.

Bleep bloop. Greetings Ben Pfaff, 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. git-am: .git/rebase-apply/patch:42: trailing whitespace. primary/

[ovs-dev] Thanks

Email me at,ra1356...@gmail.com,with your full names for more information about the unclaim fund I discovered. ___ dev mailing list d...@openvswitch.org https://mail.openvswitch.org/mailman/listinfo/ovs-dev

[ovs-dev] お支払い方法の情報を更新

お支払い方法の情報を更新してください。Update default card for your membership. マイストア? タイムセール? ギフト券 Amazonプライムをご利用頂きありがとうございます。お客様のAmazonプライム会員資格は、2020/07/08に更新を迎えます。お調べしたところ、会費のお支払いに使用できる有効なクレジットカードがアカウントに登録されていません。クレジットカード情報の更新、新しいクレジットカードの追加については以下の手順をご確認ください。 1. アカウントサービスからAmazonプライム会員情報を管理するにアクセスします。

Re: [ovs-dev] [PATCH v9] AB bonding: Add "primary" interface concept

Jeff Squyres writes: > In AB bonding, if the current active slave becomes disabled, a > replacement slave is arbitrarily picked from the remaining set of > enabled slaves. This commit adds the concept of a "primary" slave: an > interface that will always be (or become) the current active slave i

Re: [ovs-dev] [PATCH v6 0/6] DPCLS Subtable ISA Optimization

On Mon, Jul 6, 2020 at 4:57 AM Van Haaren, Harry wrote: > > > -Original Message- > > From: William Tu > > Sent: Sunday, July 5, 2020 3:06 PM > > To: Van Haaren, Harry > > Cc: ovs-dev ; Stokes, Ian ; > > Ilya > > Maximets ; Federico Iezzi > > Subject: Re: [PATCH v6 0/6] DPCLS Subtable I

Re: [ovs-dev] [PATCH V5 11/12] netdev-offload-dpdk: Support offload of clone tnl_push/output actions

On 7/7/20 1:54 PM, Eli Britstein wrote: > Tunnel encapsulation is done by tnl_push and output actions nested in a > clone action. Support offloading of such flows with > RTE_FLOW_ACTION_TYPE_RAW_ENCAP attribute. > > Signed-off-by: Eli Britstein > Reviewed-by: Oz Shlomo > Acked-by: Sriharsha Basa

Re: [ovs-dev] [PATCH v9] AB bonding: Add "primary" interface concept

On Jul 2, 2020, at 4:17 PM, Flavio Leitner wrote: > > > Hi Jeff, > > Thanks for following up. > I have a couple comments inline. > > > On Thu, Jun 25, 2020 at 02:03:34PM -0700, Jeff Squyres via dev wrote: >> In AB bonding, if the current active slave becomes disabled, a >> replacement slave i

Re: [ovs-dev] OVS Soft Freeze date for 2.14 release

On 7/3/20 7:56 AM, Simon Horman wrote: > Hi Maximets, > > On Thu, Jul 02, 2020 at 08:28:21PM +0200, Ilya Maximets wrote: >> Hi Ben, >> >> We reached the "soft freeze" date yesterday. So, we probably need >> to make announcement on a mailing list that soft freeze is started. >> >> And we need to an

[ovs-dev] OVS "soft freeze" is in effect.

Hi. As described in Documentation/internals/release-process.rst, we are in a "soft freeze" state: During the freeze, we ask committers to refrain from applying patches that add new features unless those patches were already being publicly discussed and reviewed before the freeze began.

[ovs-dev] お支払い方法の情報を更新

お支払い方法の情報を更新してください。Update default card for your membership. マイストア?タイムセール?ギフト券 Amazonプライムをご利用頂きありがとうございます。お客様のAmazonプライム会員資格は、2020/06/09に更新を迎えます。お調べしたところ、会費のお支払いに使用できる有効なクレジットカードがアカウントに登録されていません。クレジットカード情報の更新、新しいクレジットカードの追加については以下の手順をご確認ください。 1. アカウントサービスからAmazonプライム会員情報を管理するにアクセスします。

[ovs-dev] [PATCH v10] AB bonding: Add "primary" interface concept

In AB bonding, if the current active slave becomes disabled, a replacement slave is arbitrarily picked from the remaining set of enabled slaves. This commit adds the concept of a "primary" slave: an interface that will always be (or become) the current active slave if it is enabled. The rationale

[ovs-dev] [PATCH] netdev-linux: Fix broken build on Ubuntu 14.04

Patch 29cf9c1b3b9c ("userspace: Add TCP Segmentation Offload support") uses __virtio16 which is defined in kernel 3.19. Ubuntu 14.04 is using 3.13 kernel that lacks the virtio_types definition. This patch fixes that. Fixes: 29cf9c1b3b9c ("userspace: Add TCP Segmentation Offload support") Signed-

[ovs-dev] Un buen plan de vida y carrera.

Buenas Tardes Quise aprovechar la oportunidad de hacerte una invitación para tomar nuestro curso: Nombre: Planes de carrera y programas de sucesión. Horario: de 10:00 am a 17:00 pm. ¿Cuándo?: Jueves 23 de Julio Formato: En línea con interacción en vivo. Lugar: En Vivo desde su computadora La i

[ovs-dev] [PATCH V6 03/12] dpif-netdev: Don't use zero flow mark

Zero flow mark is used to indicate the HW to remove the mark. A packet marked with zero mark is received in SW without a mark at all, so it cannot be used as a valid mark. Change the pool range to fix it. Fixes: 241bad15d99a ("dpif-netdev: associate flow with a mark id") Signed-off-by: Eli Britste

[ovs-dev] [PATCH V6 01/12] netdev-offload-dpdk: Log testpmd format for flow create/destroy

To enhance debugability with DPDK, format the logs in a testpmd format commands. Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai --- lib/netdev-offload-dpdk.c | 380 -- 1 file changed, 167 insertions(+), 213 deletions(-) diff --git a/lib/net

[ovs-dev] [PATCH V6 02/12] dpif-netdev: Add mega ufid in flow add/del log

As offload is done using the mega ufid of a flow, for better debugability, add it in the log message. Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai --- lib/dpif-netdev.c | 13 + tests/dpif-netdev.at| 2 ++ tests/ofproto-macros.at | 3 ++- 3 files changed, 13 i

[ovs-dev] [PATCH V6 06/12] netdev-offload-dpdk: Remove pre-validate of patterns function

The function of adding patterns by requested matches checks that it consumed all the required matches, and err if not. For functional purpose there is no need for pre-validation. For performance such validation may decrease the time spent for failing flows, but at the expense of increasing the time

[ovs-dev] [PATCH V6 05/12] netdev-offload-dpdk: Support partial TCP/UDP port matching

The cited commit failed partial matching of TCP/UDP port matching, preventing such offload of supporting HWs. Remove this failure. Fixes: e8a2b5bf92bb ("netdev-dpdk: implement flow offload with rte flow") Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai Acked-by: Sriharsha Basavapatna -

[ovs-dev] [PATCH V6 08/12] netdev-offload-dpdk: Support offload of set IPv6 actions

Add support for set IPv6 actions. Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai Acked-by: Sriharsha Basavapatna --- Documentation/howto/dpdk.rst | 1 + NEWS | 2 ++ lib/netdev-offload-dpdk.c| 32 3 files changed, 35 inse

[ovs-dev] [PATCH V6 12/12] netdev-offload-dpdk: Support tnl/push using vxlan encap attribute

For DPDK, there is the RAW_ENCAP attribute which gets raw buffer of the encapsulation header. For specific protocol, such as vxlan, there is a more specific attribute, VXLAN_ENCAP, which gets the parsed fields of the outer header. In case tunnel type is vxlan, parse the header and use the specific

[ovs-dev] [PATCH V6 00/12] netdev datapath offload: Support IPv6 and VXLAN encap

This patch set includes enhanced logging to increase debugability, bug fixes and additional offloads - IPv6 and VXLAN encap. Patches #1-#2: Enhance log prints for debugability. Patches #3-#4: Bug fixes for partial offloads and Ethernet matching. Patches #5-#12: Add support for offloads of IPv6 p

[ovs-dev] [PATCH V6 04/12] netdev-offload-dpdk: Fix Ethernet matching for type only

For OVS rule of the form "eth type is 0x1234 / end", rule is offloaded in the form of "eth / end", which is incorrect. Fix it. Fixes: e8a2b5bf92bb ("netdev-dpdk: implement flow offload with rte flow") Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai Acked-by: Sriharsha Basavapatna ---

[ovs-dev] [PATCH V6 10/12] netdev-offload: Use dpif type instead of class.

From: Ilya Maximets There is no real difference between the 'class' and 'type' in the context of common lookup operations inside netdev-offload module because it only checks the value of pointers without using the value itself. However, 'type' has some meaning and can be used by offload provides

[ovs-dev] [PATCH V6 09/12] netdev: Allow storing dpif type into netdev structure.

From: Ilya Maximets Storing of the dpif type of the owning datapath interface will allow us to easily distinguish, for example, userspace tunneling ports from the system ones. This is required in terms of HW offloading to avoid offloading of userspace flows to kernel interfaces that doesn't belo

[ovs-dev] [PATCH V6 11/12] netdev-offload-dpdk: Support offload of clone tnl_push/output actions

Tunnel encapsulation is done by tnl_push and output actions nested in a clone action. Support offloading of such flows with RTE_FLOW_ACTION_TYPE_RAW_ENCAP attribute. Signed-off-by: Eli Britstein Reviewed-by: Oz Shlomo Acked-by: Sriharsha Basavapatna --- Documentation/howto/dpdk.rst | 1 + NEW

[ovs-dev] [PATCH V6 07/12] netdev-offload-dpdk: Add IPv6 pattern matching

Add support for IPv6 pattern matching for offloading flows. Signed-off-by: Eli Britstein Reviewed-by: Roni Bar Yanai Acked-by: Sriharsha Basavapatna --- Documentation/howto/dpdk.rst | 2 +- NEWS | 1 + lib/netdev-offload-dpdk.c| 75