Re: [ovs-dev] [PATCH v3] netdev-dummy: Limits the number of tx/rx queues.

2017-01-09 21:56 GMT-08:00 nickcooper-zhangtonghao : > This patch avoids the ovs_rcu to report WARN, caused by blocked > for a long time, when ovs-vswitchd processes a port with many > rx/tx queues. The number of tx/rx queues per port may be appropriate, > because the dpdk

[ovs-dev] Ref:PO71887/Awaiting for draft contract letter/PI

Hello dear, Happy new year to you and your entire family. We attached our first orders for the year of 2017 and you have to make sure the items marked on yellow are to be treated urgently and include them with the first shipment. Please kindly make a draft of the contract letter alongside

Re: [ovs-dev] intermittent ovs-vswitchd segfault

On Wed, Jan 11, 2017 at 5:24 AM, Joe Stringer wrote: > On 10 January 2017 at 00:25, Numan Siddique wrote: > > Hi, > > > > I am seeing intermittent segfault's in ovs-vswitchd. We have like 20 > > compute nodes and noticed the crash in 4 or 5 nodes. Seems to me

Re: [ovs-dev] *** SPAM *** [PATCH] xlate: Recirculate also when MPLS POP is implicit.

> On Jan 6, 2017, at 9:06 AM, Ben Pfaff wrote: > > On Thu, Jan 05, 2017 at 06:19:41PM -0800, Jarno Rajahalme wrote: >> 'ctx->was_mpls' is used to flag when an MPLS packet has been popped to >> a non-MPLS packet, but it was not set when the MPLS POP is implicit >> due to the

Re: [ovs-dev] [PATCH] python/setup: "-fPIC" extra compile flag for C extension

Hi, On 2017年01月11日 11:01, Ben Pfaff wrote: > On Wed, Jan 11, 2017 at 09:38:13AM +0900, IWASE Yusuke wrote: >> To build shared library object, it is required to built as "Position >> Independent Code" (on x86-64, for example). >> This patch adds "-fPIC" extra compile flag to make sure that

Re: [ovs-dev] [PATCH] python/setup: "-fPIC" extra compile flag for C extension

On Wed, Jan 11, 2017 at 09:38:13AM +0900, IWASE Yusuke wrote: > To build shared library object, it is required to built as "Position > Independent Code" (on x86-64, for example). > This patch adds "-fPIC" extra compile flag to make sure that extension > being PIC while building a shared library

[ovs-dev] (no subject)

Brauchen Sie ein Geschäft oder persönliches Darlehen. Wir geben Darlehen im Bereich von € 1.000 bis € 100.000.000 EURO, KONTAKTIEREN SIE UNS JETZT ÜBER EMAIL: jerryloa...@gmail.com ___ dev mailing list d...@openvswitch.org

Re: [ovs-dev] [PATCH v3] dpcls: Avoid one 8-byte chunk in subtable mask.

Thanks for the update! Pushed to master with small edits for clarification. Jarno > On Jan 6, 2017, at 1:45 AM, antonio.fische...@intel.com wrote: > > This patch allows to skip the chunk comprising of dp_hash and in_port > in the subtable mask when the packet is not recirculated. This will

[ovs-dev] [PATCH] python/setup: "-fPIC" extra compile flag for C extension

To build shared library object, it is required to built as "Position Independent Code" (on x86-64, for example). This patch adds "-fPIC" extra compile flag to make sure that extension being PIC while building a shared library object. Signed-off-by: IWASE Yusuke ---

Re: [ovs-dev] [PATCH 2/2] revalidator: Complain for more ukey transitions.

In the worst case this patch will introduce rate-limited logging for other states than UKEY_OPERATIONAL that may want to transition onto themselves, so: Acked-by: Jarno Rajahalme > On Jan 10, 2017, at 3:54 PM, Joe Stringer wrote: > > For most ukey transition

Re: [ovs-dev] [PATCH 1/2] revalidator: Prevent double-delete of ukey.

Acked-by: Jarno Rajahalme > On Jan 10, 2017, at 3:54 PM, Joe Stringer wrote: > > revalidator_sweep__() splits checking for whether to delete a ukey from > the actual deletion to prevent taking the umap lock for too long. > However it uses information gathered from

Re: [ovs-dev] intermittent ovs-vswitchd segfault

On 10 January 2017 at 00:25, Numan Siddique wrote: > Hi, > > I am seeing intermittent segfault's in ovs-vswitchd. We have like 20 > compute nodes and noticed the crash in 4 or 5 nodes. Seems to me the crash > is seen when the system is idle for a long time (as I noticed over

[ovs-dev] [PATCH] Windows: Implement Hyper-V VIF discovery agent.

Signed-off-by: Yin Lin --- windows/OvsDiscoveryAgent/App.config | 18 ++ windows/OvsDiscoveryAgent/OvsDiscoveryAgent.csproj | 83 windows/OvsDiscoveryAgent/OvsDiscoveryAgent.sln| 34 +++ windows/OvsDiscoveryAgent/OvsDiscoveryService.cs | 67

Re: [ovs-dev] [PATCH] ovn: Support ARP proxy in logical switches.

Han, Thanks - that makes sense to me. I can’t claim that this is a hugely general feature, but your use case seems legit to me (with the caveat that I’m no k8s expert). Bruce On Jan 9, 2017, at 4:08 PM, Han Zhou > wrote: Hi Bruce, This feature is

Re: [ovs-dev] [PATCH ovs V2 12/21] dpif-netlink: Use netdev flow put api to insert a flow

On 10 January 2017 at 07:50, Paul Blakey wrote: >>> +nl_msg_put_u32(, OVS_ACTION_ATTR_OUTPUT, ifindex_out); >>> +if (tnl_cfg && tnl_cfg->dst_port != 0) { >>> +nl_msg_put_u32(, OVS_TUNNEL_KEY_ATTR_TP_DST, >>> tnl_cfg->dst_port); >>> +

Re: [ovs-dev] [PATCH ovs V2 12/21] dpif-netlink: Use netdev flow put api to insert a flow

On 10 January 2017 at 06:36, Paul Blakey wrote: > > > On 06/01/2017 01:28, Joe Stringer wrote: >> >> On 25 December 2016 at 03:39, Paul Blakey wrote: >>> >>> Using the new netdev flow api operate will now try and >>> offload flows to the relevant netdev of

Re: [ovs-dev] [PATCH ovs V2 09/21] dpif-netlink: Dump netdevs flows on flow dump

On 10 January 2017 at 03:45, Paul Blakey wrote: >>> +struct netdev_list_element *element; >>> +struct ovs_list port_list; >>> +int ports = netdev_hmap_port_get_list(dpif_->dpif_class, >>> _list); >>> +int i = 0; >>> + >>> +dump->netdev_dumps = >>> +

Re: [ovs-dev] [PATCH] ovsdb-data: Add support for integer ranges in database commands

Thanks! Please let me know if you think I can help implementing something or be useful in some other way. I'll also try to propose something. 2017-01-05 17:48 GMT+01:00 Ben Pfaff : > On Thu, Dec 29, 2016 at 03:55:46PM -0700, Lukasz Rzasik wrote: > > Adding / removing a range of

Re: [ovs-dev] [PATCH ovs V2 05/21] dpif: Save added ports in a port map for netdev flow api use

On 10 January 2017 at 03:20, Paul Blakey wrote: > > > On 05/01/2017 03:54, Joe Stringer wrote: >> >> On 25 December 2016 at 03:39, Paul Blakey wrote: >>> >>> To use netdev flow offloading api, dpifs needs to iterate over >>> added ports. This addition

Re: [ovs-dev] [PATCH v3 0/4] Fix some "clone"-related issues

I applied these to master following Dong Jun's testing. On Fri, Jan 06, 2017 at 08:33:41AM -0800, Ben Pfaff wrote: > v1->v2: > - Patches 1 and 2 were applied and dropped. > - Patch 3 (now patch 1) no longer clears the action set or stack, although > it > saves and restores them,

Re: [ovs-dev] Issues with the use of the clone action for resubmission to the pipeline

Thank you very much for testing. I applied all four patches to master. On Sat, Jan 07, 2017 at 06:08:15PM +0800, Dong Jun wrote: > > On 2017/1/7 17:14, Dong Jun wrote: > >I tested my > >issue(https://mail.openvswitch.org/pipermail/ovs-dev/2016-December/326936.html) > >withpatch serial v3 1-4

Re: [ovs-dev] [PATCH] windows: Change driver and MSI company name to LF

Acked-by: Sairam Venugopal On 12/6/16, 3:30 PM, "ovs-dev-boun...@openvswitch.org on behalf of Alin Serdean" wrote: >Until now we used 'Open vSwitch' as the company/organization name. > >The

[ovs-dev] USA Global Youths Conference 2017

Dear Sir / Madam, It is my great privilege for me to invite you to 2017 Global Youths conferences which will be held from 21st - 25th of February 2017 at the {Atlanta Georgia Convention Center United States of America, The organizing committee is responsible for all visa arrangements to the

Re: [ovs-dev] [PATCH ovs V2 00/21] Introducing HW offload support for openvswitch

On 10 January 2017 at 10:07, Joe Stringer wrote: > On 8 January 2017 at 07:18, Paul Blakey wrote: >> >> On 05/01/2017 02:11, Joe Stringer wrote: >> >>> On 25 December 2016 at 03:39, Paul Blakey wrote: This patch series introduces

Re: [ovs-dev] [PATCH v2 4/5] datapath-windows: Add support for OVS_KEY_ATTR_UDP set action

Udp checksum should not be updated in the case it was 0. I will add the changes in the next version. Thanks, Alin. > -Original Message- > From: Alin Serdean > Sent: Friday, January 6, 2017 9:34 PM > To: d...@openvswitch.org > Cc: Alin Serdean > Subject:

[ovs-dev] [PATCH 4/6] datapath-windows: VXLAN Check for flow destination port

Change the UDP destination port(VXLAN header) to check if it was set by the userspace, use it if it was set. If the userspace did not specify a destination port, use the configured vport destination port. Signed-off-by: Alin Gabriel Serdean ---

[ovs-dev] [PATCH 3/6] datapath-windows: Add support for OVS_TUNNEL_KEY_ATTR_TP_DST

Add support for netlink attribute OVS_TUNNEL_KEY_ATTR_TP_DST get/set flow functions. Signed-off-by: Alin Gabriel Serdean --- datapath-windows/ovsext/Flow.c | 8 1 file changed, 8 insertions(+) diff --git a/datapath-windows/ovsext/Flow.c

[ovs-dev] [PATCH 2/6] datapath-windows: Allow tunnel action to modify destination port

'OvsTunnelAttrToIPv4TunnelKey' modifies 'tunkey' with the received netlink attributes(i.e. OVS_TUNNEL_KEY_ATTR_IPV4_DST). Change the order of the value assignment to reflect the values received via userspace. Signed-off-by: Alin Gabriel Serdean ---

[ovs-dev] [PATCH 1/6] datapath-windows: Fix alignment in MapTunAttrToFlowPut

Found by inspection. Signed-off-by: Alin Gabriel Serdean --- datapath-windows/ovsext/Flow.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/datapath-windows/ovsext/Flow.c b/datapath-windows/ovsext/Flow.c index 2e8b42b..749d83a 100644 ---

[ovs-dev] [PATCH] debian; Avoid installing ovs-vswitchd.conf.db manpage as "db" language.

Usually, when the name of a manpage has a two-letter extension, it means that the manpage is written in the language designated by that language code. Reported-by: Michael Stapelberg Reported-at: https://bugs.debian.org/850631 Signed-off-by: Ben Pfaff ---

Re: [ovs-dev] [PATCH ovs V2 01/21] tc: Add tc flower interface

On 05/01/2017 02:57, Joe Stringer wrote: On 25 December 2016 at 03:39, Paul Blakey wrote: Add tc flower interface that will be used to offload flows via tc flower classifier. Depending on the flag used (skip_sw/hw) flower will pass those to HW or handle them itself.

Re: [ovs-dev] [PATCH ovs V2 19/21] dpctl: read vswitch config on start

On 05/01/2017 23:46, Joe Stringer wrote: On 25 December 2016 at 03:39, Paul Blakey wrote: Use Open vSwitch IDL pattern to read OVS configuration on dpctl start, needed as some functionality is dependent on that configuration. Signed-off-by: Paul Blakey

[ovs-dev] Bug#850631: marked as done (Manpage ovs-vswitchd.conf.db.5 installed into wrong directory)

Your message dated Tue, 10 Jan 2017 16:34:13 + with message-id and subject line Bug#850631: fixed in openvswitch 2.6.2~pre+git20161223-3 has caused the Debian Bug report #850631, regarding Manpage ovs-vswitchd.conf.db.5 installed into wrong directory to be

Re: [ovs-dev] [PATCH ovs V2 12/21] dpif-netlink: Use netdev flow put api to insert a flow

On 06/01/2017 01:25, Joe Stringer wrote: On 25 December 2016 at 03:39, Paul Blakey wrote: Using the new netdev flow api operate will now try and offload flows to the relevant netdev of the input port. Other operate methods flows will come in later patches. Signed-off-by:

Re: [ovs-dev] [PATCH] ovn-ctl: Modify SYNC FROM connection default protocol to SSL

> From: "e" > To: ovs-dev@openvswitch.org > Cc: "e" > Sent: Monday, January 9, 2017 9:44:43 PM > Subject: [ovs-dev] [PATCH] ovn-ctl: Modify SYNC FROM connection default > protocol to SSL > > This patch is used for the OVSDB HA by pacemaker. > which

Re: [ovs-dev] [RFC PATCH] datapath: allow tunnels to be created with rtnetlink

On Tue, Jan 10, 2017 at 04:26:10PM +0100, Jiri Benc wrote: > On Tue, 10 Jan 2017 15:44:07 +0100, Jiri Benc wrote: > > create an ovs_geneve interface using rtnetlink > > This can be done just once. The pseudocode at startup thus would be: > > create an ovs_geneve interface using rtnetlink > if

[ovs-dev] [RFC] libX: add new release / version info tags

This commit uses the $PACKAGE_VERSION automake variable to construct a release and version info combination which sets the library name to be: libfoo-$(OVS_MAJOR_VERSION).so.$(OVS_MINOR_VERSION).0.$(OVS_MICRO_VERSION) where formerly, it was always: libfoo.so.1.0.0 This allows releases of

Re: [ovs-dev] [RFC PATCH] datapath: allow tunnels to be created with rtnetlink

On Tue, Jan 10, 2017 at 8:14 PM, Jiri Benc wrote: > On Tue, 10 Jan 2017 19:29:21 +0530, Pravin Shelar wrote: >> OVS kernel module has compile time checks for various kernel features, >> if any of required tunnel feature is missing OVS kernel module >> compiles in support for its

Re: [ovs-dev] [RFC PATCH] datapath: allow tunnels to be created with rtnetlink

On Tue, 10 Jan 2017 15:44:07 +0100, Jiri Benc wrote: > create an ovs_geneve interface using rtnetlink This can be done just once. The pseudocode at startup thus would be: create an ovs_geneve interface using rtnetlink if successful { delete the created interface set out_of_tree

Re: [ovs-dev] [PATCH ovs V2 12/21] dpif-netlink: Use netdev flow put api to insert a flow

On 06/01/2017 01:28, Joe Stringer wrote: On 25 December 2016 at 03:39, Paul Blakey wrote: Using the new netdev flow api operate will now try and offload flows to the relevant netdev of the input port. Other operate methods flows will come in later patches. Signed-off-by:

Re: [ovs-dev] [RFC PATCH] datapath: allow tunnels to be created with rtnetlink

On Tue, Jan 10, 2017 at 3:23 AM, Eric Garver wrote: > On Fri, Dec 09, 2016 at 03:12:39PM -0800, Pravin Shelar wrote: >> On Fri, Dec 9, 2016 at 12:43 AM, Jiri Benc wrote: >> > On Thu, 8 Dec 2016 22:49:56 -0800, Pravin Shelar wrote: >> >> OVS out of tree kernel

Re: [ovs-dev] [PATCH ovs V2 14/21] netdev-tc-offloads: Netdev flow put implementation using tc api

On 05/01/2017 23:29, Joe Stringer wrote: On 25 December 2016 at 03:39, Paul Blakey wrote: Signed-off-by: Paul Blakey Reviewed-by: Roi Dayan --- I don't think that there are errors returned for unsupported fields? Any field that

Re: [ovs-dev] [PATCH ovs V2 09/21] dpif-netlink: Dump netdevs flows on flow dump

On 05/01/2017 23:27, Joe Stringer wrote: On 25 December 2016 at 03:39, Paul Blakey wrote: While dumping flows, dump flows that were offloaded to netdev and parse them back to dpif flow. Signed-off-by: Paul Blakey Reviewed-by: Roi Dayan

Re: [ovs-dev] [PATCH ovs V2 05/21] dpif: Save added ports in a port map for netdev flow api use

On 05/01/2017 03:54, Joe Stringer wrote: On 25 December 2016 at 03:39, Paul Blakey wrote: To use netdev flow offloading api, dpifs needs to iterate over added ports. This addition inserts the added dpif ports in a hash map, The map will also be used to translate dpif

Re: [ovs-dev] [PATCH v2 5/5] datapath-windows: Add support for OVS_KEY_ATTR_TCP set action

Hi Sai, Thanks for the review! Comments inlined. > -Original Message- > From: Sairam Venugopal [mailto:vsai...@vmware.com] > Sent: Tuesday, January 10, 2017 3:40 AM > To: Alin Serdean ; > d...@openvswitch.org > Subject: Re: [ovs-dev] [PATCH v2 5/5]