[ovs-dev] [PATCH] github: Add Appveyor image and link to Readme

Just a small nit to see current build status of appveyor. Also add a link so one could easily reach the history of the builds. Signed-off-by: Alin Gabriel Serdean --- README.rst | 2 ++ 1 file changed, 2 insertions(+) diff --git a/README.rst b/README.rst index

[ovs-dev] [PATCH 4/9] ofproto/trace: Query ct_state for conntrack recirc from DP

Instead of using fixed default conntrack state 'trk|new' in ofproto/trace for conntrack recirculation, this patch queries the conntrack state from datapath using ct_dpif_get_info(). Signed-off-by: Yi-Hung Wei --- lib/ct-dpif.c| 42

[ovs-dev] [PATCH 5/9] dpif-netlink: Implement ct_get_info

This patch implements ct_get_info() in dpif-netlink. It uses NFNL_SUBSYS_CTNETLINK netlink subsystem to query conntrack info from kernel datapath. Then, ofproto/trace can use the ct_get_info() to derive the ct_state of the traced flow. System traffic tests are provided to verify that ofproto/trace

[ovs-dev] [PATCH 6/9] netlink-conntrack: Add support for querying conntrack exp

The ct_state of an uncommmited new flow is marked as related if the flow is in the conntrack expectation table. In order for ofproto/trace to identify the ct_state of a new related flow, this patch utilizes NFNL_SUBSYS_CTNETLINK_EXP netlink subsystem to query the conntrack expectation table,

[ovs-dev] [PATCH 7/9] flow: Refactor parse_ct_state()

Refactor parse_ct_state() to support different delimiters. Signed-off-by: Yi-Hung Wei --- lib/flow.c | 6 +++--- lib/flow.h | 2 +- ofproto/ofproto-dpif-trace.c | 2 +- ovn/utilities/ovn-trace.c| 2 +- 4 files changed, 6

Re: [ovs-dev] [RFC PATCH] daemon-windows: Set default error mode for services

Thanks for the patch. Acked-by: Anand Kumar Regards, Anand Kumar On 8/25/17, 10:48 AM, "ovs-dev-boun...@openvswitch.org on behalf of Alin Gabriel Serdean" wrote: Microsoft allows default process

Re: [ovs-dev] [PATCH net-next v6 2/3] net: gso: Add GSO support for NSH

On Fri, 25 Aug 2017 18:25:14 +0200, Jiri Benc wrote: > While looking at this, I realized that GSO for VXLAN-GPE is broken, > too. Let me fix it by implementing what I described above which will > make your patch much easier. Okay, it's not really broken and we don't need that complexity. At least

Re: [ovs-dev] [PATCH v4] dpif-netdev: Avoid reading RSS hash when EMC is disabled

> -Original Message- > From: Darrell Ball [mailto:db...@vmware.com] > Sent: Friday, August 25, 2017 6:19 PM > To: Fischetti, Antonio ; d...@openvswitch.org > Subject: Re: [ovs-dev] [PATCH v4] dpif-netdev: Avoid reading RSS hash when EMC > is disabled > > Hi

[ovs-dev] [PATCH 2/9] ofproto/trace: Propagate ct_zone in recirculation

This patch propagates ct_zone when ofproto/trace automatically runs through the recirculation process. Fixes: e6bc8e749381 ("ofproto/trace: Add support for tracing conntrack recirculation") Signed-off-by: Yi-Hung Wei --- ofproto/ofproto-dpif-trace.c | 4 +++-

[ovs-dev] [PATCH 3/9] ct-dpif: Add ct_dpif_get_info()

This patch adds ct_dpif_get_info() to dpif_class for querying conntrack info from datapath. Later patches will use this function to query ct_fields such as ct_state, ct_mark, and ct_label for ofproto/trace command. The following commits will provide implementation of ct_dpif_get_info() on

[ovs-dev] [PATCH 9/9] ofproto/trace: Support ct_mark and ct_label in --ct-next

Previously, --ct-next option in ofproto/trace only supports specifying ct_state. This patch adds support of ct_mark and ct_label. Signed-off-by: Yi-Hung Wei --- lib/ct-dpif.c| 7 lib/ct-dpif.h| 3 ++ lib/netlink-conntrack.c | 5

[ovs-dev] [PATCH 8/9] ofproto/trace: Change delimiter of ct_state in --ct-next

This patch changes the delimiter of ct_state in --ct-next from comma or space to vertical bar. The new format will be easier to parse when the --ct-next options support more ct_fields such as ct_mark and ct_label. Signed-off-by: Yi-Hung Wei --- ofproto/ofproto-dpif-trace.c

Re: [ovs-dev] OVS-DPDK public meeting

23rd August 2017 ATTENDEES: Aaron, Ben, Flavio, Darrell, Antonio, Ian, Georg, Michael, Ori, Rohit, Sammeh, Yippeng, Frikkie, Kevin, Billy, Zoltan (Missed some) === GENERAL === - OVS 2.8 -- Finalizing NSH commits for 2.8 -- Aiming for release next week -- RH testing (Flavio) --- Found

[ovs-dev] [PATCH v4] netdev-dpdk: Create separate memory pool for each port.

Since it's possible to delete memory pool in DPDK we can try to estimate better required memory size when port is reconfigured, e.g. with different number of rx queues. CC: Kevin Traynor CC: Aaron Conole Signed-off-by: Robert Wojciechowicz

[ovs-dev] [PATCH] dpif: Remove duplicated word in comment for dpif_recv()

From: Kaige Fu Signed-off-by: Kaige Fu --- lib/dpif.c | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/lib/dpif.c b/lib/dpif.c index 79b2e6c..a4471d5 100644 --- a/lib/dpif.c +++ b/lib/dpif.c @@ -1547,11 +1547,11 @@

Re: [ovs-dev] [PATCH v4] dpif-netdev: Avoid reading RSS hash when EMC is disabled

On 8/25/17, 3:21 PM, "Fischetti, Antonio" wrote: > -Original Message- > From: Darrell Ball [mailto:db...@vmware.com] > Sent: Friday, August 25, 2017 6:19 PM > To: Fischetti, Antonio ;

Re: [ovs-dev] [PATCH v2] conntrack: Fix ct-clean thread crash bug

Thanks for the fix Lili I also updated AUTHORS.rst, since I did not find your name there. I also added a Fixes tag and more detail to the commit message. and applied the patch to dpdk_merge here

[ovs-dev] [PATCH 0/2] vHost Dequeue Zero Copy

This patch enables optional dequeue zero copy for vHost ports. This gives a performance increase for some use cases. I'm using the cover letter to report my results. vhost (vm1) -> vhost (vm2) Using testpmd to source (txonly) in vm1 and sink (rxonly) in vm2. 4C1Q 64B packets: 5.05Mpps -> 5.52Mpps

Re: [ovs-dev] [PATCH v3 2/4] dpif-netdev: Avoid reading RSS hash when EMC is disabled

Sure Darrell, I'll do that. Thanks! Antonio > -Original Message- > From: Darrell Ball [mailto:db...@vmware.com] > Sent: Friday, August 25, 2017 9:59 AM > To: Fischetti, Antonio ; d...@openvswitch.org > Subject: Re: [ovs-dev] [PATCH v3 2/4] dpif-netdev: Avoid

Re: [ovs-dev] [PATCH] netdev-dpdk: vhost get stats fix

OK, I can separate this. Regarding this issue, my idea: I donot think it is libvirtd bug. We can have different stats set, say common set, extended set and so on, but we should NOT have one paramter level difference, since it is too hard to stats user to parse it. Now I think it is better

[ovs-dev] [PATCH] netdev-dpdk: vhost get stats fix.

Signed-off-by: wangzhike --- lib/netdev-dpdk.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c index 1aaf6f7..bf5c58b 100644 --- a/lib/netdev-dpdk.c +++ b/lib/netdev-dpdk.c @@ -2016,10 +2016,10 @@

Re: [ovs-dev] [PATCH v3 0/4] Skip EMC for recirc pkts and other optimizations.

I applied patches 3 and 4 to https://github.com/darball/ovs/commits/dpdk_merge with minor rewording of one sentence. Patch 2 is presently dependent on Patch 1, although it could be split out I think and Patch 1 is still being discussed. On 8/11/17, 8:52 AM, "ovs-dev-boun...@openvswitch.org on

Re: [ovs-dev] [PATCH] netdev-dpdk: vhost get stats fix

I am wondering if we should split the +stats->tx_errors = 0; out from this patch and discuss it separately ? In theory, if a stat is really not supported, we should not display a value for it. Displaying 0 could be misleading if there really is a problem and we are not detecting it.

Re: [ovs-dev] [PATCH v3 2/4] dpif-netdev: Avoid reading RSS hash when EMC is disabled

Hi Antonio Can the dependency of this Patch 2 on Patch 1 be removed, while Patch 1 is being discussed ? Thanks Darrell On 8/13/17, 11:32 PM, "ovs-dev-boun...@openvswitch.org on behalf of Darrell Ball" wrote: I did not try

Re: [ovs-dev] [PATCH] conntrack: Fix ct-clean thread crash bug

Yes, it works fine to me. And I will update the [PATCH v2]. Thanks for the advice. - On 8/24/17, 11:03 AM, "Darrell Ball" wrote: Thanks for testing. I’ll look at in detail

[ovs-dev] [PATCH v2] conntrack: Fix ct-clean thread crash bug

From: Lili Huang Conn should be removed from the list before freed. This crash will be triggered when a established flow do ct(nat) again, like "ip,actions=ct(table=1) table=1,in_port=1,ip,actions=ct(commit,nat(dst=5.5.5.5)),2

Re: [ovs-dev] ICT HELP DESK

Verdachte e-mails worden verzonden vanuit dit account. Het wordt daarom uitgeschakeld vanuit de systeemdatabase, die u niet meer verzendt en ontvangt. Voor Mail account recovery klik HIER en volg de instructies. Opmerking: Niet-valideren,

Re: [ovs-dev] [PATCH v6 0/6] OVS-DPDK rxq to pmd assignment improvements.

I applied the series to https://github.com/darball/ovs/commits/dpdk_merge On 8/24/17, 4:37 PM, "Kevin Traynor" wrote: For the DPDK datapath, by default rxqs are assigned to available pmds in round robin order with no weight or priority. It can happen that

Re: [ovs-dev] [RFC patch v1] dp-packet: Refactor DPDK packet initialization.

Thanks for the review Sugesh I added your Acked-by I applied it to https://github.com/darball/ovs/commits/dpdk_merge On 8/18/17, 2:32 AM, "Chandran, Sugesh" wrote: Can you release the patch for this? Looks to me it’s a good to have. Regards

[ovs-dev] [PATCH 1/2] windows, python: create a different event for sockets

At the moment the sockets on Windows use the same events that are being created for the pipes. This is not correct because they should be different events. This patch introduces a new event which should be used for sockets. The new event needs to be set on automatic reset with its initial state

[ovs-dev] [PATCH net-next v6 0/3] openvswitch: add NSH support

v5->v6 - Fix the rest comments for v4. - Add NSH GSO support for VxLAN-gpe + NSH and Eth + NSH. v4->v5 - Fix many comments by Jiri Benc and Eric Garver for v4. v3->v4 - Add new NSH match field ttl - Update NSH header to the latest format which will be final format and won't change

[ovs-dev] [PATCH v4] dpif-netdev: Avoid reading RSS hash when EMC is disabled

When EMC is disabled the reading of RSS hash is skipped. Also, for packets that are not recirculated it retrieves the hash value without considering the recirc id. Signed-off-by: Antonio Fischetti --- V4 - reworked to remove dependencies from other patches in

[ovs-dev] [PATCH net-next v6 1/3] net: add NSH header structures and helpers

NSH (Network Service Header)[1] is a new protocol for service function chaining, it can be handled as a L3 protocol like IPv4 and IPv6, Eth + NSH + Inner packet or VxLAN-gpe + NSH + Inner packet are two typical use cases. This patch adds NSH header structures and helpers for NSH GSO support and

[ovs-dev] [PATCH net-next v6 2/3] net: gso: Add GSO support for NSH

NSH (Network Service Header)[1] is a new protocol for service function chaining, it can be handled as a L3 protocol like IPv4 and IPv6, Eth + NSH + Inner packet or VxLAN-gpe + NSH + Inner packet are two typical use cases. We need to enbale Open vSwitch to support NSH, this patch is to make Linux

[ovs-dev] [PATCH net-next v6 3/3] openvswitch: enable NSH support

OVS master and 2.8 branch has merged NSH userspace patch series, this patch is to enable NSH support in kernel data path in order that OVS can support NSH in compat mode by porting this. Signed-off-by: Yi Yang --- drivers/net/vxlan.c | 7 +

[ovs-dev] [PATCH] windows,python: remove dead code in send_wait

Found while looking over the code. Signed-off-by: Alin Balutoiu Suggested-by: Alin Gabriel Serdean --- python/ovs/stream.py | 2 -- 1 file changed, 2 deletions(-) diff --git a/python/ovs/stream.py b/python/ovs/stream.py index

Re: [ovs-dev] [PATCH 00/40] Fix static code analysis warnings.

> -Original Message- > From: Shashank Ram [mailto:r...@vmware.com] > Sent: Friday, August 25, 2017 2:56 AM > To: Ben Pfaff ; Alin Serdean > > Cc: d...@openvswitch.org > Subject: Re: [ovs-dev] [PATCH 00/40] Fix static code analysis warnings.

Re: [ovs-dev] [PATCH net-next v6 1/3] net: add NSH header structures and helpers

On Fri, 25 Aug 2017 22:20:03 +0800, Yi Yang wrote: > --- a/include/uapi/linux/if_ether.h > +++ b/include/uapi/linux/if_ether.h > @@ -138,6 +138,7 @@ > #define ETH_P_IEEE802154 0x00F6 /* IEEE802.15.4 frame > */ > #define ETH_P_CAIF 0x00F7 /* ST-Ericsson CAIF

Re: [ovs-dev] [PATCH v2] ovn: Check for known logical switch port types.

On Thu, Aug 24, 2017 at 2:42 PM Russell Bryant wrote: > On Thu, Aug 24, 2017 at 2:13 PM, Mark Michelson > wrote: > > OVN is lenient with the types of logical switch ports. Maybe too > > lenient. This patch attempts to solve this problem on two fronts: > > >

Re: [ovs-dev] [PATCH 0/7 RFC] OVS-DPDK flow offload with rte_flow

Thank you for this contribution It is a big step forward for HWOL; even partial offload for these patches have a significant benefit. I had done a first pass on Wednesday when the patches came out but I’ll send comments out on Monday On 8/22/17, 11:24 PM, "Yuanhan Liu"

Re: [ovs-dev] [PATCH v4] dpif-netdev: Avoid reading RSS hash when EMC is disabled

Hi Antonio On 8/25/17, 6:56 AM, "ovs-dev-boun...@openvswitch.org on behalf of antonio.fische...@intel.com" wrote: When EMC is disabled the reading of RSS hash is skipped. Also, for packets that are not

[ovs-dev] [RFC PATCH] daemon-windows: Set default error mode for services

Microsoft allows default process memory dumps via WER (Windows Error Reporting). WER can be set to collect dumps using in general using: https://msdn.microsoft.com/en-us/library/windows/desktop/bb787181(v=vs.85).aspx If a normal application crashes, the user will receive a pop-up dialog, in which

[ovs-dev] [RFC PATCH v2 01/10] acinclude.m4: Support compilation of libIPsec.

LibIpsecMB is required to enable the use of vdev cryptodev devices in DPDK. This patch adds a condition to check for the library when it is detected that ONFIG_RTE_LIBRTE_PMD_AESNI_MB=y is enabled in the DPDK config. Signed-off-by: Ian Stokes --- acinclude.m4 | 13

[ovs-dev] [RFC PATCH v2 04/10] flow: Add ESP spi value to flow struct.

This patch adds a field to the flow struct to represent the ESP security parameter index of a packet. Signed-off-by: Ian Stokes --- include/openvswitch/flow.h |5 + 1 files changed, 5 insertions(+), 0 deletions(-) diff --git a/include/openvswitch/flow.h

[ovs-dev] [RFC PATCH v2 03/10] packets: Add ESP header and trailer.

This patch introduces structs for both ESP headers and ESP trailers along with expected size assertions. Signed-off-by: Ian Stokes --- lib/packets.h | 14 ++ 1 files changed, 14 insertions(+), 0 deletions(-) diff --git a/lib/packets.h b/lib/packets.h index

[ovs-dev] [RFC PATCH v2 08/10] vswitch.xml: Detail vxlanipsec user interface.

This commit adds details to the vswitch xml regarding the use of the vxlanipsec interface type. This patch is not intended for upstreaming and simply seeks to solicit feedback on the user interface design of the vxlanipsec port type as described in the vswitch.xml. This modifies the vswitch.xml

Re: [ovs-dev] [PATCH net-next v6 2/3] net: gso: Add GSO support for NSH

On Fri, 25 Aug 2017 22:20:04 +0800, Yi Yang wrote: > --- a/include/linux/skbuff.h > +++ b/include/linux/skbuff.h > @@ -766,7 +766,7 @@ struct sk_buff { > __u8ndisc_nodetype:2; > #endif > __u8ipvs_property:1; > - __u8

[ovs-dev] [RFC PATCH v2 00/10] RFC for Userspace IPsec Interface

Hi all, I've started to work on enabling IPsec for userspace in OVS. This RFC patchset provides a very basic implementation allowing users to test a simple VM to VM setup. The patch set is far from complete and will require a lot more work before I consider submitting from upstream but I'm

[ovs-dev] [RFC PATCH v2 02/10] openvswitch.h: add vport to ovs_action_push_tnl.

Upon callback for building/pushing a packet header when encapsulating for tunneling a reference to the vport in question is required to access associated devices such as cryptodevs. This patch adds this pointer and will enable future work with cryptodevs that are associated with a vport.

[ovs-dev] [RFC PATCH v2 09/10] Docs: Add userspace-ipsec how to guide.

This commit adds a how to guide for using the proposed vxlanipsec userspace interface. It is not intended to be upstreamed but simply seeks to solicit feed back by providing an example of the proposed vxlanipsec interface design setup and usage. The how to usecase deals with securing vxlan

[ovs-dev] [PATCH v3] ovn: Check for known logical switch port types.

OVN is lenient with the types of logical switch ports. Maybe too lenient. This patch attempts to solve this problem on two fronts: 1) In ovn-nbctl, if you attempt to set the port type to an unknown type, the command will not end up setting the type. 2) In northd, when copying the port type from