[ovs-dev] [PATCH] ovn: Fix a northd bug.

The 'chassis' member in the 'struct sbrec_binding' must always be non-null. However, this is not case when creating the binding in "set_bindings()". And it causes segfault while starting northd with existing ovnnb configuration. This commit fixes the bug by always setting the 'chassis' to an emp

[ovs-dev] [PATCH] ovs-ofctl: Always prints recirc_id in decimal

The output of 'ovs-ofctl dump-flows' command prints recirc_id in decimal in action parts of the output, while prints that in hex in matching parts of the same output. This patch fixes the inconsistency by always printing recirc_id values in decimal. Reported-by: Justin Pettit Signed-off-by: Andy

Re: [ovs-dev] [PATCH v2] ovn-controller: Add physical/logical translation for containers.

> > In this kind of situation I'd tend to just write 'tag ? 150 : 100' > for the priority. I made the change and pushed it. Thanks for the review! ___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev

Re: [ovs-dev] OVS 2.3.1 build fail using openwrt buildroot

I don't know; this is the first report of this kind that I remember seeing. On Mon, May 18, 2015 at 02:28:53PM -0400, Alison C. wrote: > Okay. Any other ideas why the build might be failing, then? > > 2015-05-15 12:27 GMT-04:00 Ben Pfaff : > > On Fri, May 15, 2015 at 11:56:42AM -0400, Alison C. w

Re: [ovs-dev] [PATCH v2] ovn-controller: Add physical/logical translation for containers.

On Mon, May 18, 2015 at 06:03:41AM -0700, Gurucharan Shetty wrote: > When containers are running inside VMs and the openflow flows > are added in the hypervisor, the physical to logical translation > (and vice versa) needs to handle the VLAN tags that the packet > comes with. > > Signed-off-by: Gu

Re: [ovs-dev] [PATCH 16/19] ovs-ofctl: Add 'bundle' command and unit testing.

> On May 15, 2015, at 11:01 AM, Jarno Rajahalme wrote: > > >> On May 15, 2015, at 10:03 AM, Romain Lenglet > > wrote: >> >> On May 15, 2015 at 7:45:31 AM, Jarno Rajahalme (jrajaha...@nicira.com >> (mailto:jrajaha...@nicira.com >

[ovs-dev] [PATCH v2 19/19] Use classifier versioning.

Each rule is now added or deleted in a specific tables version. Flow tables are versioned with a monotonically increasing 64-bit integer, where positive values are valid version numbers. Rule modifications are implemented as an insertion of a new rule and a deletion of the old rule, both taking p

[ovs-dev] [PATCH v2 18/19] ofproto: Infra for table versioning.

Signed-off-by: Jarno Rajahalme --- ofproto/ofproto-dpif-xlate.c | 17 - ofproto/ofproto-dpif.c | 43 +- ofproto/ofproto-dpif.h |5 - ofproto/ofproto-provider.h |6 ++ ofproto/ofproto.c|5

[ovs-dev] [PATCH v2 17/19] classifier: Support table versioning

This patch allows classifier rules to become visible and invisible in specific versions. A 'version' is defined as a positive monotonically increasing integer, which never wraps around. When versioning is not used, the 'version' parameter should be passed as 'CLS_NO_VERSION'. This feature enable

[ovs-dev] [PATCH v2 15/19] ofproto: Add support for reverting flow mods, and bundle commit.

Signed-off-by: Jarno Rajahalme --- ofproto/bundles.c | 40 - ofproto/bundles.h | 21 +++-- ofproto/ofproto.c | 235 +++-- tests/ofproto.at | 66 +++ 4 files changed, 293 insertions(+), 69 deletions(-) diff --git a/ofprot

[ovs-dev] [PATCH v2 16/19] ovs-ofctl: Add 'bundle' command and unit testing.

The new ovs-ofctl 'bundle' command accepts files similar to 'add-flows', but each line can optionally start with 'add', 'modify', 'delete', 'modify_strict', or 'delete_strict' keyword, so that arbitrary flow table modifications may be specified in a single file. The modifications in the file are ex

[ovs-dev] [PATCH v2 14/19] ofproto: Split delete_flow*().

Signed-off-by: Jarno Rajahalme --- ofproto/ofproto.c | 79 +++-- 1 file changed, 65 insertions(+), 14 deletions(-) diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c index fca3f2c..2d5507c 100644 --- a/ofproto/ofproto.c +++ b/ofproto/ofproto.c @@

[ovs-dev] [PATCH v2 12/19] ofproto: Split add_flow().

Split add_flow() to add_flow_begin() which does all the error checking, and add_flow_finish() which can not fail. Since we still want to send an error response for an unknown 'buffer_id', send_buffered_packet() now send the error response itself. Signed-off-by: Jarno Rajahalme --- ofproto/ofpro

[ovs-dev] [PATCH v2 11/19] ofproto: Refactor modify_flows__().

Signed-off-by: Jarno Rajahalme --- ofproto/ofproto.c | 317 +++-- 1 file changed, 187 insertions(+), 130 deletions(-) diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c index 6a8239e..5c8b1a5 100644 --- a/ofproto/ofproto.c +++ b/ofproto/ofproto.c

[ovs-dev] [PATCH v2 13/19] ofproto: Split modify_flows().

Signed-off-by: Jarno Rajahalme --- ofproto/ofproto.c | 114 - 1 file changed, 78 insertions(+), 36 deletions(-) diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c index d3fce08..fca3f2c 100644 --- a/ofproto/ofproto.c +++ b/ofproto/ofproto.c @@

[ovs-dev] [PATCH v2 08/19] ofproto: Factor out ofproto_rule_insert__().

Pairs well with ofproto_rule_remove__(). Signed-off-by: Jarno Rajahalme --- ofproto/ofproto.c | 29 + 1 file changed, 21 insertions(+), 8 deletions(-) diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c index 54d66b6..d9f2249 100644 --- a/ofproto/ofproto.c +++ b/ofp

[ovs-dev] [PATCH v2 09/19] ofproto: Factor out ofproto_rule_create().

Makes add_flow() easier to read and understand. Signed-off-by: Jarno Rajahalme --- ofproto/ofproto.c | 96 - 1 file changed, 59 insertions(+), 37 deletions(-) diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c index d9f2249..d0124e0 100644 -

[ovs-dev] [PATCH v2 10/19] ofproto: Use enum ofperr for 'error'.

Signed-off-by: Jarno Rajahalme --- ofproto/ofproto.c | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c index d0124e0..6a8239e 100644 --- a/ofproto/ofproto.c +++ b/ofproto/ofproto.c @@ -3866,7 +3866,7 @@ collect_rules_strict(st

[ovs-dev] [PATCH v2 07/19] bundles: Manage bundles in connmgr.

This will make implementing commit in ofproto.c easier. Signed-off-by: Jarno Rajahalme --- ofproto/bundles.c | 116 ++--- ofproto/bundles.h | 19 - ofproto/connmgr.c | 51 +-- ofproto/connmgr.h |6 ++- 4 files c

[ovs-dev] [PATCH v2 06/19] bundles: Validate bundled messages.

OpenFlow bundle messages should be decoded and validated at the time they are added to the bundle. This commit does this for flow mod and port mod messages. Signed-off-by: Jarno Rajahalme --- lib/ofp-print.c |4 ++-- lib/ofp-util.c| 10 +++--- lib/ofp-util.h|7 +-- o

[ovs-dev] [PATCH v2 05/19] classifier: Support duplicate rules.

OpenFlow 1.4 bundles are easier to implement when it is possible to mark a rule as 'to_be_removed' and then insert a new, identical rule with the same priority. All but one out of the identical rules must be marked as 'to_be_removed', and the one rule that is not 'to_be_removed' must have been ins

[ovs-dev] [PATCH v2 04/19] classifier: Add support for invisible flows.

This makes it possible to tentatively add flows to the classifier without the datapath seeing them. Signed-off-by: Jarno Rajahalme --- lib/classifier-private.h |3 +- lib/classifier.c | 68 +- lib/classifier.h |1 + ofproto/of

[ovs-dev] [PATCH v2 03/19] ofp-util: Fix xid in ofputil_encode_bundle_add().

Bundle add must use the same xid as the embedded message. Signed-off-by: Jarno Rajahalme --- lib/ofp-util.c |4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/lib/ofp-util.c b/lib/ofp-util.c index 6366919..e62c584 100644 --- a/lib/ofp-util.c +++ b/lib/ofp-util.c @@ -8900,7

[ovs-dev] [PATCH v2 02/19] ofp-parse: Remove unused F_OUT_PORT field flag.

Signed-off-by: Jarno Rajahalme --- lib/ofp-parse.c |7 +++ 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/lib/ofp-parse.c b/lib/ofp-parse.c index 856044d..126980c 100644 --- a/lib/ofp-parse.c +++ b/lib/ofp-parse.c @@ -245,7 +245,6 @@ parse_ofp_str__(struct ofputil_flow_mod

[ovs-dev] [PATCH v2 01/19] ofproto: Eliminate use of unset error code.

Signed-off-by: Jarno Rajahalme --- ofproto/ofproto.c |3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c index e062732..5c3b497 100644 --- a/ofproto/ofproto.c +++ b/ofproto/ofproto.c @@ -4372,8 +4372,7 @@ add_flow(struct ofproto *ofproto,

[ovs-dev] [PATCH v2 00/19] Support OF bundles for flow mods.

This series completes the implementation of OpenFlow 1.4 bundles for flow mods. Port mods are not yet supported. OpenFlow 1.4 specifies support for 'bundles' which allow arbitrary collections of flow table modifications to be executed as a single transaction. Supporting this in OVS requres major

Re: [ovs-dev] [PATCH v2 0/7] Userspace datapath performance improvements

On Mon, May 18, 2015 at 10:47 AM, Daniele Di Proietto wrote: > This series contains different tweaks to improve the performance of the > userspace datapath with DPDK ports. > > The first commits reduce the size of struct dp_packet to three cachelines > (two used by DPDK and one for our metadata).

Re: [ovs-dev] I am thinking about f%cking you

Send me a F#ckFriends request so we can hook up My usename is GetMeOffSamantha27 http://totalglass.eu/dnxs8k/3etf8.html";>my profile is here CANT WAIT ___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev

[ovs-dev] [PATCH v2] ovn-controller: Add physical/logical translation for containers.

When containers are running inside VMs and the openflow flows are added in the hypervisor, the physical to logical translation (and vice versa) needs to handle the VLAN tags that the packet comes with. Signed-off-by: Gurucharan Shetty --- v1->v2: * For container traffic that are tagged, strip the

Re: [ovs-dev] dev Digest, Vol 70, Issue 91

    h Sent from my BlackBerry 10 smartphone.   Original Message   From: dev-requ...@openvswitch.org Sent: Monday, May 18, 2015 1:50 PM To: dev@openvswitch.org Reply To: dev@openvswitch.org Subject: dev Digest, Vol 70, Issue 91 Send dev mailing list submissions to dev@openvswitch.org To subscribe

[ovs-dev] [PATCH v5] netdev-dpdk: add dpdk vhost-user ports

This patch adds support for a new port type to the userspace datapath called dpdkvhostuser. A new dpdkvhostuser port will create a unix domain socket which when provided to QEMU is used to facilitate communication between the virtio-net device on the VM and the OVS port on the host. vhost-cuse ('

Re: [ovs-dev] [PATCH] datapath-windows: Fix warning from the powershell module

Acked-by: Eitan Eliahu Thanks Alin. Eitan -Original Message- From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Alin Serdean Sent: Monday, May 04, 2015 9:45 AM To: dev@openvswitch.org Subject: [ovs-dev] [PATCH] datapath-windows: Fix warning from the powershell module This patc

Re: [ovs-dev] [PATCH] ARP lookup and next hop functionality on windows

Hi Alin, It looks good besides some minor comments. Thanks, Eitan -Original Message- From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Alin Serdean Sent: Monday, May 04, 2015 8:35 AM To: dev@openvswitch.org Subject: [ovs-dev] [PATCH] ARP lookup and next hop functionality on wind

Re: [ovs-dev] [PATCH ovn] ovn-controller: Add physical/logical translation for containers.

On Mon, May 18, 2015 at 07:32:01AM -0700, Gurucharan Shetty wrote: > > > >> +int tag = 0; > >> +ofp_port_t ofport; > >> +if (binding->parent_port) { > >> +ofport = u16_to_ofp(simap_get(&lport_to_ofport, > >> + binding->par

Re: [ovs-dev] OVS 2.3.1 build fail using openwrt buildroot

Okay. Any other ideas why the build might be failing, then? 2015-05-15 12:27 GMT-04:00 Ben Pfaff : > On Fri, May 15, 2015 at 11:56:42AM -0400, Alison C. wrote: >> Hey everybody, >> >> Is there a minimum binutils version required to build OVS 2.3.1? >> >> I'm trying to build it in a openwrt buildro

Re: [ovs-dev] [PATCH] ofproto: Fix memory leak in flow deletion.

Pushed to master, thanks for the review! I checked branch-2.3, no backports are needed. Jarno > On May 16, 2015, at 10:07 PM, Ben Pfaff wrote: > > On Fri, May 15, 2015 at 10:59:41AM -0700, Jarno Rajahalme wrote: >> Fix a memory leak that was introduced in commit 834fe5cb997b (ofproto: >> Add

[ovs-dev] [PATCH v2 1/7] dp-packet: Remove 'list' member.

The 'list' member is only used (two users) in the slow path. This commit removes it to reduce the struct size Signed-off-by: Daniele Di Proietto --- lib/dp-packet.c| 13 - lib/dp-packet.h| 8 lib/netdev-dummy.c | 48 ++

[ovs-dev] [PATCH v2 4/7] dp-packet: Style fixes.

Also, removes an unused function Signed-off-by: Daniele Di Proietto --- lib/dp-packet.h | 230 1 file changed, 131 insertions(+), 99 deletions(-) diff --git a/lib/dp-packet.h b/lib/dp-packet.h index 1d10d99..e4c2593 100644 --- a/lib/dp-pa

Re: [ovs-dev] netdev-dpdk: Doubt about rings in dpdkr port type

On Mon, May 18, 2015 at 7:09 PM, Traynor, Kevin wrote: > > > -Original Message- > > From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Mauricio > Vásquez > > Sent: Monday, May 18, 2015 9:31 AM > > To: dev@openvswitch.org; ivano cerrato > > Subject: [ovs-dev] netdev-dpdk: Doubt ab

[ovs-dev] [PATCH v2 7/7] dpif-netdev: Share emc and fast path output batches.

Until now the exact match cache processing was able to handle only four megaflows. The rest of the packets was passed to the megaflow classifier. The limit was arbitraly set to four also because the algorithm used to group packets in output batches didn't perform well with a lot of megaflows. Af

[ovs-dev] [PATCH v2 6/7] dpif-netdev: Store batch pointer in dp_netdev_flow.

The userspace datapath 1. receives a batch of packets. 2. finds a 'netdev_flow' (megaflow) for each packet. 3. groups the packets in output batches based on the 'netdev_flow'. Until now the grouping (2) was done using a simple algorithm with a O(N^2) runtime, where N is the number of distinct meg

Re: [ovs-dev] [PATCH 7/7] dpif-netdev: Share emc and fast path output batches.

On 15/05/2015 20:51, "Pravin Shelar" wrote: >On Thu, Apr 23, 2015 at 11:40 AM, Daniele Di Proietto > wrote: >> Until now the exact match cache processing was able to handle only four >> megaflow. The rest of the packets was passed to the megaflow >> classifier. >> >> The limit was arbitraly se

[ovs-dev] [PATCH v2 5/7] dpif-netdev: Store pkt_metadata structure in dp_netdev_port.

Initializing a struct pkt_metadata for every packet can be surprisingly expensive. It's much faster to keep a copy for each port and copying it on each packet. Suggested-by: Pravin Shelar Signed-off-by: Daniele Di Proietto --- lib/dpif-netdev.c | 23 --- 1 file changed, 12

[ovs-dev] [PATCH v2 3/7] dp-packet: Merge 'allocated' member with DPDK mbuf 'buf_len'.

DPDK buf_len is only 16-bit wide ('allocated' was 32-bit), but it should be enough to store the number of allocated bytes. This will reduce 'struct dp_packet' size. Signed-off-by: Daniele Di Proietto --- lib/dp-packet.c | 6 +++--- lib/dp-packet.h | 26 +++--- 2 files chang

[ovs-dev] [PATCH v2 2/7] dp-packet: Remove 'frame' member.

In 'struct ofpbuf' the 'frame' pointer was used to parse different kinds of data (Ethernet, OpenFlow, Netlink attributes). For Ethernet packets the 'frame' pointer was supposed to have the same value as the 'data' pointer. Since 'struct dp_packet' is only used for Ethernet packets, there's no nee

[ovs-dev] [PATCH v2 0/7] Userspace datapath performance improvements

This series contains different tweaks to improve the performance of the userspace datapath with DPDK ports. The first commits reduce the size of struct dp_packet to three cachelines (two used by DPDK and one for our metadata). I've put in also some style fixes for lib/dp-packet.h Then, a microopt

Re: [ovs-dev] [PATCH 5/7] dpif-netdev: Use memcpy() to initialize pkt_metadata.

On 15/05/2015 20:50, "Pravin Shelar" wrote: >On Thu, Apr 23, 2015 at 11:40 AM, Daniele Di Proietto > wrote: >> Initializing the dp_packet's metadata can be a hot spot, especially >> for very simple pipelines. Therefore improving the code here can >> sometimes make a difference. >> >> Using mem

Re: [ovs-dev] [PATCH 1/7] dp-packet: Remove 'list' member.

On 15/05/2015 20:50, "Pravin Shelar" wrote: >On Thu, Apr 23, 2015 at 11:39 AM, Daniele Di Proietto > wrote: >> The 'list' member is only used (two users) in the slow path. >> This commit removes it to reduce the struct size >> >> Signed-off-by: Daniele Di Proietto >> --- >> lib/dp-packet.c

Re: [ovs-dev] [PATCH 1/2] sparse: Fix sparse when compiling DPDK.

On my system (I'm using GCC 4.9 and sparse 0.4.5) I need lots of other headers like emmintrin.h to compile without warnings. The complete list is: adxintrin.h,ammintrin.h,avx2intrin.h,avx512fintrin.h, avxintrin.h,bmi2intrin.h,bmiintrin.h,f16cintrin.h, fma4intrin.h,fmaintrin.h,fxsrintrin.h,ia32int

Re: [ovs-dev] [PATCH 2/2] netdev-dpdk: Fix sparse warnings.

Acked-by: Daniele Di Proietto On 18/05/2015 16:53, "Ethan Jackson" wrote: >These are all minor style issues. > >Signed-off-by: Ethan Jackson >--- > lib/netdev-dpdk.c | 22 ++ > 1 file changed, 14 insertions(+), 8 deletions(-) > >diff --git a/lib/netdev-dpdk.c b/lib/netdev-d

Re: [ovs-dev] [PATCH] dpdk: Ditch MAX_PKT_BURST macro.

> -Original Message- > From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Ethan Jackson > Sent: Monday, May 18, 2015 5:08 PM > To: dev@openvswitch.org > Subject: Re: [ovs-dev] [PATCH] dpdk: Ditch MAX_PKT_BURST macro. > > This version of the patch breaks sparse, I sent out another

Re: [ovs-dev] netdev-dpdk: Doubt about rings in dpdkr port type

> -Original Message- > From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Mauricio Vásquez > Sent: Monday, May 18, 2015 9:31 AM > To: dev@openvswitch.org; ivano cerrato > Subject: [ovs-dev] netdev-dpdk: Doubt about rings in dpdkr port type > > I'm performing a series of VM2VM com

[ovs-dev] [PATCH] Extend the sFlow agent to report tunnel and MPLS structures

Packets are still sampled at ingress only, so the egress tunnel and/or MPLS structures are only included when there is just 1 output port. The actions are either provided by the datapath in the sample upcall or looked up in the userspace cache. The former is preferred because it is more reliable

Re: [ovs-dev] [PATCH] dpdk: Ditch MAX_PKT_BURST macro.

Thanks for taking care of this! There are some "NETDEV_MAX_RX_BATCH" left in dpif-netdev.c (inside some #else blocks). Changing those: Acked-by: Daniele Di Proietto On 16/05/2015 16:24, "Ethan Jackson" wrote: >The MAX_PKT_BURST and NETDEV_MAX_RX_BATCH macros had a confusing >relationship. T

Re: [ovs-dev] [PATCH] dpdk: Ditch MAX_PKT_BURST macro.

This version of the patch breaks sparse, I sent out another. Ethan On Sat, May 16, 2015 at 11:24 AM, Ethan Jackson wrote: > The MAX_PKT_BURST and NETDEV_MAX_RX_BATCH macros had a confusing > relationship. They basically purport to do the same thing, making it > unclear which is the source of tr

[ovs-dev] [PATCH v2] dpdk: Ditch MAX_PKT_BURST macro.

The MAX_PKT_BURST and NETDEV_MAX_RX_BATCH macros had a confusing relationship. They basically purport to do the same thing, making it unclear which is the source of truth. Furthermore, while NETDEV_MAX_RX_BATCH was 256, MAX_PKT_BURST was 32, meaning we never process a batch larger than 32 packets

[ovs-dev] [PATCH 2/2] netdev-dpdk: Fix sparse warnings.

These are all minor style issues. Signed-off-by: Ethan Jackson --- lib/netdev-dpdk.c | 22 ++ 1 file changed, 14 insertions(+), 8 deletions(-) diff --git a/lib/netdev-dpdk.c b/lib/netdev-dpdk.c index 505ab75..5f8c60f 100644 --- a/lib/netdev-dpdk.c +++ b/lib/netdev-dpdk.c @@

[ovs-dev] [PATCH 1/2] sparse: Fix sparse when compiling DPDK.

Sparse doesn't like several of the DPDK header files. This patch works around it so we can get analysis when compiling DPDK. Signed-off-by: Ethan Jackson --- include/sparse/automake.mk | 4 include/sparse/emmintrin.h | 21 + include/sparse/rte_atomic.h | 25 +

Re: [ovs-dev] [PATCH v4] netdev-dpdk: add dpdk vhost-user ports

On 05/13/2015 04:15 PM, Ciara Loftus wrote: This patch adds support for a new port type to the userspace datapath called dpdkvhostuser. A new dpdkvhostuser port will create a unix domain socket which when provided to QEMU is used to facilitate communication between the virtio-net device on the V

Re: [ovs-dev] [PATCH ovn] ovn-controller: Add physical/logical translation for containers.

> >> +int tag = 0; >> +ofp_port_t ofport; >> +if (binding->parent_port) { >> +ofport = u16_to_ofp(simap_get(&lport_to_ofport, >> + binding->parent_port)); >> +if (ofport) { > > I don't think that it is safe to

Re: [ovs-dev] [PATCH 5/5] datapath-windows: Multiple NBLs support for VXLAN packets

Hi Nithin, I agree with you. We can drop this patch then. Sorin -Original Message- From: Nithin Raju [mailto:nit...@vmware.com] Sent: Thursday, 14 May, 2015 18:24 To: Sorin Vinturis Cc: dev@openvswitch.org Subject: Re: [ovs-dev] [PATCH 5/5] datapath-windows: Multiple NBLs support for V

[ovs-dev] netdev-dpdk: Doubt about rings in dpdkr port type

I'm performing a series of VM2VM communication testing where a virtual machine is a packet generator while the other one just receives all the packets. When I tested dpdk-ovs[1] I used a port type called dpdkclient: this port has 4 rings, (tx, rx, alloc_q and free_q), alloc_q and free_q are used b