Re: [ovs-dev] [RFC] [PATCH 00/11] Data Path Classifier Offloading

Thank you for the contribution and RFC status acknowledged. On 7/5/17, 9:32 PM, "ovs-dev-boun...@openvswitch.org on behalf of Shachar Beiser" wrote: Hi , I would like to clarify since I did not

Re: [ovs-dev] [RFC] [PATCH 00/11] Data Path Classifier Offloading

Hi , I would like to clarify since I did not add the label [RFC] to the subject of my patches , but only to the cover letter that All the patches I have sent were sent as a RFC: [RFC] [PATCH 00/11] Data Path Classifier Offloading [RFC][PATCH 1/11] ovs/dp-cls [RFC][PATCH 2/11]

[ovs-dev] [patch_v4 7/8] System tests: Add a few passive ftp tests.

In order to have full coverage of ALGs for the userspace datapath, it is necessary to add 4 new tests. Three of these will cover passive ftp, including basic V6 passive ftp, V4 passive ftp with NAT and sequence skew and V6 passive ftp with NAT. The last test will cover tftp with NAT. Before these

[ovs-dev] [patch_v4 8/8] NEWS: Announce userspace datapath ALG support.

Signed-off-by: Darrell Ball --- NEWS | 1 + 1 file changed, 1 insertion(+) diff --git a/NEWS b/NEWS index 0f2604f..fc5c8f4 100644 --- a/NEWS +++ b/NEWS @@ -29,6 +29,7 @@ Post-v2.7.0 abbreviated to 4 hex digits. - Userspace Datapath: * Added NAT support for

[ovs-dev] [patch_v4 4/8] Userspace Datapath: Add ALG infra and FTP.

ALG infra and FTP (both V4 and V6) support is added to the userspace datapath. Also, NAT support is included. Signed-off-by: Darrell Ball --- lib/conntrack-private.h | 20 +- lib/conntrack.c | 1027 +++ lib/conntrack.h

[ovs-dev] [patch_v4 6/8] System tests: Enable ALGs for userspace.

Signed-off-by: Darrell Ball --- tests/system-userspace-macros.at | 7 ++- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/tests/system-userspace-macros.at b/tests/system-userspace-macros.at index 6e3d468..417aa4f 100644 --- a/tests/system-userspace-macros.at

[ovs-dev] [patch_v4 5/8] Userspace Datapath: Add TFTP support.

Both ipv4 and ipv6 are supported. Also, NAT support is included. Signed-off-by: Darrell Ball --- include/sparse/netinet/in.h | 1 + lib/conntrack.c | 45 - 2 files changed, 45 insertions(+), 1 deletion(-) diff --git

[ovs-dev] [patch_v4 2/8] Userspace Datapath: Remove 'nat' from lock name.

The 'nat' portion of 'nat_resources_lock' is dropped as this lock will be used by ALGs in a subsequent patch. Signed-off-by: Darrell Ball --- lib/conntrack.c | 30 +++--- lib/conntrack.h | 2 +- 2 files changed, 16 insertions(+), 16 deletions(-) diff

[ovs-dev] [patch_v4 3/8] Userspace Datapath: Add missed lock annotations.

Fixes: a489b16854b5 ("conntrack: New userspace connection tracker.") Fixes: 286de2729955 ("dpdk: Userspace Datapath: Introduce NAT Support.") Signed-off-by: Darrell Ball --- lib/conntrack.c | 4 1 file changed, 4 insertions(+) diff --git a/lib/conntrack.c

[ovs-dev] [patch_v4 1/8] Userspace Datapath: Change 'related' to 'icmp_related'.

The conntrack context flag 'related' is changed to 'icmp_related' to disambiguate usage w.r.t. ALGs which are added in a subsequent patch. Signed-off-by: Darrell Ball --- lib/conntrack.c | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git

[ovs-dev] [patch_v4 0/8] Userspace Datapath: Add ALG support.

ALG infra is added with support for FTP and TFTP. Both V4 and V6 are supported. Also, NAT is supported. Three passive ftp system tests are added to complete testing coverage of ftp for the userspace datapath, as the existing coverage of passive ftp was limited to one part of one test for V4

Re: [ovs-dev] [PATCH] checkpatch: Use default encoding from email library.

On Tue, Jul 04, 2017 at 07:16:46AM -0700, Joe Stringer wrote: > There are three paths for running the core checkpatch path: From a file, > from stdin, or reading from git output. Currently, the file version of > this calls the "email" library's decode routine which translates the > stream into a

Re: [ovs-dev] offset of 0x10001 in dpif_netlink_queue_to_priority

On Tue, Jul 04, 2017 at 05:09:21PM +0200, Matthias May wrote: > Hi > > I'm trying to map the vlan_pcp onto 802.11 tid. > > net/wireless/util.c in mac80211 specifies: > > > /* skb->priority values from 256->263 are magic values to > > * directly indicate a specific 802.1d priority.

[ovs-dev] [PATCH] Support IPv6 link-local address scopes on Linux.

I hadn't even heard of this feature before, but it seems to be at least semi-standard to support Linux link-local address scopes via a % suffix, e.g. fe80::1234%eth0 for a link-local address scoped to eth0. This commit adds support. I'd appreciate feedback from folks who understand this feature

Re: [ovs-dev] [PATCHv2 2/2] ofp-parse: Improve comment for parse_ofp_packet_out_str().

On Tue, Jun 27, 2017 at 06:14:44PM -0700, Justin Pettit wrote: > This function is a wrapper for parse_ofp_packet_out_str__(), which > states that a couple members must be freed on success. This commit > updates the wrapper's description to indicate the same. > > The existing callers appear to be

Re: [ovs-dev] [PATCHv2 1/2] ofp-parse: Fix small memory leak when calling parse_ofp_meter_mod_str().

On Tue, Jun 27, 2017 at 06:14:43PM -0700, Justin Pettit wrote: > The function parse_ofp_meter_mod_str() allocates a buffer called > 'bands', which parse_ofp_meter_mod_str__() then steals for the member > 'mm->meter.bands'. Calling functions didn't free that stolen value and > the comments for

Re: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release.

On Wed, Jul 05, 2017 at 03:48:01PM +, Stokes, Ian wrote: > > >From: ovs-dev-boun...@openvswitch.org > > >[mailto:ovs-dev-boun...@openvswitch.org] > > >On Behalf Of Ian Stokes > > >Sent: Tuesday, June 20, 2017 1:57 PM > > >To: d...@openvswitch.org > > >Subject: [ovs-dev] [PATCH v1] docs: Use

Re: [ovs-dev] [PATCH] ofproto-dpif-ipfix: Fix severe memory leak in ipfix_send_template_msgs().

On Tue, Jun 27, 2017 at 11:54:08PM -0700, Justin Pettit wrote: > > > On Jun 1, 2017, at 8:46 PM, Ben Pfaff wrote: > > > > On Thu, Jun 01, 2017 at 05:11:37PM -0700, Justin Pettit wrote: > >> > >>> On May 26, 2017, at 9:14 PM, Ben Pfaff wrote: > >>> > >>> This fixes

Re: [ovs-dev] [PATCH v2 2/4] redhat: dynamically allocate and reference ovs user

On 07/05/2017 10:16 PM, Aaron Conole wrote: > Hi Markos, > [...] >> >> I am a bit puzzled about this to be honest... I am wondering if it would >> be better to do it the other way around. For example, supply a sysconfig >> file with OVS_USER_ID commented out, but if it's an upgrade, then do the >>

Re: [ovs-dev] [PATCH] ovn-controller: use localnet port for directly connected datapath only.

On Wed, Jul 5, 2017 at 9:07 AM, Mickey Spiegel wrote: > > > On Wed, Jul 5, 2017 at 12:36 AM, Han Zhou wrote: >> >> >> >> On Tue, Jul 4, 2017 at 10:56 PM, Mickey Spiegel wrote: >> > >> > >> > On Tue, Jun 27, 2017 at 10:42 AM, Han

[ovs-dev] [PATCH] [RPM] Update openvswitch-kmod-fedora.spec

Update the kernel module spec file to better reflect how I'm locally building the kmod RPM in mock and pushing it to COPR. These changes also better align with RPM linting and uses the preferred macros for referring to the build root. Adds in some Requires and BuildRequires lines that define the

Re: [ovs-dev] [PATCH v2 2/4] redhat: dynamically allocate and reference ovs user

Hi Markos, Markos Chandras writes: > Hi Aaron, > > On 07/05/2017 08:56 PM, Aaron Conole wrote: >> After this commit, the fedora RPM will create the openvswitch user, from the >> non-static pool, for use as an Open vSwitch daemon user. This only happens >> on install - not

Re: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release.

Hi All, I've rebased the patch to specifically apply to the current 2.7 branch. I've added your acks to the commit. https://mail.openvswitch.org/pipermail/ovs-dev/2017-July/334952.html Thanks Ian > -Original Message- > From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev- >

[ovs-dev] [PATCH branch-2.7] docs: Use DPDK 16.11.2 stable release.

Modify docs and travis linux build script to use DPDK 16.11.2 stable branch to benefit from most recent bug fixes. Signed-off-by: Ian Stokes Acked-by: Mark Kavanagh Acked-by: Darrell Ball --- .travis/linux-build.sh

Re: [ovs-dev] [PATCH v2 2/4] redhat: dynamically allocate and reference ovs user

Hi Aaron, On 07/05/2017 08:56 PM, Aaron Conole wrote: > After this commit, the fedora RPM will create the openvswitch user, from the > non-static pool, for use as an Open vSwitch daemon user. This only happens > on install - not upgrade. This will be the default user:group > combination for the

Re: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release.

> -Original Message- > From: Darrell Ball [mailto:db...@vmware.com] > Sent: Wednesday, July 5, 2017 5:48 PM > To: Kavanagh, Mark B ; Stokes, Ian > ; d...@openvswitch.org > Cc: Justin Pettit > Subject: Re: [ovs-dev] [PATCH

[ovs-dev] [PATCH v2 2/4] redhat: dynamically allocate and reference ovs user

After this commit, the fedora RPM will create the openvswitch user, from the non-static pool, for use as an Open vSwitch daemon user. This only happens on install - not upgrade. This will be the default user:group combination for the openvswitch daemons. Signed-off-by: Aaron Conole

[ovs-dev] [PATCH v2 4/4] redhat: allow dpdk to also run as non-root user

After this commit, users may start a dpdk-enabled ovs setup as a non-root user. This is accomplished by exporting the $HOME directory, which dpdk uses to fill in it's semi-persistent RTE configuration. This change may be a bit controversial since it modifies /dev/hugepages as part of starting

[ovs-dev] [PATCH v2 3/4] dpdkstrip: add a preprocessor tool for stripping dpdk blocks

Normally, in C code, pre-processing macros can be used to enable/disable specific functionality based on switches passed to configure. This works for DPDK using the --with-dpdk flag, which sets the DPDK_NETDEV define to the appropriate value. However, not all files are processed with the C

[ovs-dev] [PATCH v2 0/4] rhel/fedora: non-root OvS out of the box

This series attempts to introduce the ability to start and use Open vSwitch 'out of the box' as a non-root user. It does this by modifying the service files to pass the recently introduced --ovs-user argument around, and by making some minor tweaks to the passwd, group, and filesystem

[ovs-dev] [PATCH v2 1/4] redhat: allow arbitrary user:group

Under rpm based distributions, the only user:group that the rhel daemons run as is 'root:root'. This is fine as a default, but as part of a security procedure, users may want to run as an alternate uid/gid. This commit adds an OVS_USER_ID environment variable for systemd, which defaults to

[ovs-dev] Hello

Normal 0 false false false FR X-NONE X-NONE

[ovs-dev] UPS issue #6183045: unable to delivery parcel

Dear Customer, UPS courier was unable to contact you for your parcel delivery. Postal label is enclosed to this e-mail. Please check the attachment! Kind regards, , UPS Support Manager. ___ dev mailing list d...@openvswitch.org

Re: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release.

On 7/5/17, 8:15 AM, "Kavanagh, Mark B" wrote: >From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-boun...@openvswitch.org] >On Behalf Of Stokes, Ian >Sent: Tuesday, July 4, 2017 10:09 AM >To: Darrell Ball ; d...@openvswitch.org

[ovs-dev] please let me know if you received them

Hello, I’ve not yet received a reply from you. I’mconcerned that you may not have received my email. Could you please let me knowif you received them? At my private email: nicolebmar...@hotmail.com Myregards,

Re: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release.

>From: Stokes, Ian >Sent: Wednesday, July 5, 2017 4:48 PM >To: Kavanagh, Mark B ; d...@openvswitch.org; Darrell >Ball >Cc: Justin Pettit >Subject: RE: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release. > >> >From:

Re: [ovs-dev] [PATCH] ovn-controller: use localnet port for directly connected datapath only.

On Wed, Jul 5, 2017 at 12:36 AM, Han Zhou wrote: > > > On Tue, Jul 4, 2017 at 10:56 PM, Mickey Spiegel > wrote: > > > > > > On Tue, Jun 27, 2017 at 10:42 AM, Han Zhou wrote: > >> > >> > >> > >> On Tue, Jun 27, 2017 at 10:40 AM, Han

Re: [ovs-dev] [RFC] [PATCH 00/11] Data Path Classifier Offloading

Hi , I would like to clarify since I did not add the label [RFC] to the subject of my patches , but only to the cover letter that All the patches I have sent were sent as a RFC: [RFC] [PATCH 00/11] Data Path Classifier Offloading [RFC][PATCH 1/11] ovs/dp-cls [RFC][PATCH 2/11]

Re: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release.

> >From: ovs-dev-boun...@openvswitch.org > >[mailto:ovs-dev-boun...@openvswitch.org] > >On Behalf Of Ian Stokes > >Sent: Tuesday, June 20, 2017 1:57 PM > >To: d...@openvswitch.org > >Subject: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release. > > > >Modify docs and travis linux build

Re: [ovs-dev] [PATCH 5/5] lib/netdev-dpdk: copy large packet to multi-segment mbufs

>From: Michael Qiu [mailto:qdy220091...@gmail.com] >Sent: Monday, June 19, 2017 6:29 AM >To: d...@openvswitch.org >Cc: Kavanagh, Mark B ; b...@ovn.org; >db...@vmware.com; Michael Qiu >Subject: [PATCH 5/5] lib/netdev-dpdk: copy large packet to

Re: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release.

>From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-boun...@openvswitch.org] >On Behalf Of Stokes, Ian >Sent: Tuesday, July 4, 2017 10:09 AM >To: Darrell Ball ; d...@openvswitch.org >Subject: Re: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release. > >> Hi Ian >> >> Do

Re: [ovs-dev] [PATCH v2] dpif-netlink: convert packet_type netlink attribute

Hi Zoltan, On Tue, Jul 04, 2017 at 09:23:17PM +, Zoltán Balogh wrote: > By introducing packet type-aware pipeline, match on ethertype was > removed when packet type is not Ethernet. As pointed out by Eric Garver, > this could have a side effect on the kernel datapath: >

Re: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release.

>From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-boun...@openvswitch.org] >On Behalf Of Ian Stokes >Sent: Tuesday, June 20, 2017 1:57 PM >To: d...@openvswitch.org >Subject: [ovs-dev] [PATCH v1] docs: Use DPDK 16.11.2 stable release. > >Modify docs and travis linux build script to use the

Re: [ovs-dev] Fail to netdev_open internal iface with error "File exists"

On 23/06/17 10:37, Eelco Chaudron wrote: On 06/23/2017 02:12 AM, Ben Pfaff wrote: On Thu, Jun 22, 2017 at 04:04:44PM -0300, Flavio Leitner wrote: On Thu, Jun 22, 2017 at 01:04:59AM +0800, Huanle Han wrote: Hi,all I get this problem with latest(dbd8112) branch-2.7 code on my Ubuntu.

Re: [ovs-dev] GTP support to OVS upstream

> > Dear Yi, > > I'm not fully updated on the current OpenFlow 1.6 specification. > > I have now created a workaround with two bridge interfaces on one machine > to forward normal traffic and only inspect GTP traffic. But indeed this > requires full decapsulation of the packet before we have any

Re: [ovs-dev] [PATCH v3 0/3] Incremental addition/deletion of PMD threads.

Hi all. IMHO, these patches was reviewed and tested enough to be merged. Does anybody have objections? Ben, Darrell, what do you think? Best regards, Ilya Maximets. On 15.06.2017 14:36, Ilya Maximets wrote: > Version 3: > * Added comment about 'static_txq_id's adjustment. > * Added

Re: [ovs-dev] [RFC PATCH v1 2/3] Docs: Add userspace-ipsec how to guide.

> I have given some comments for previous patch that has been addressed in > this. > So please ignore them accordingly. > > > Regards > _Sugesh > > > > -Original Message- > > From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev- > > boun...@openvswitch.org] On Behalf Of Ian Stokes > >

Re: [ovs-dev] [PATCH 4/5] lib/dp-packet: copy multi-segments data from DPDK mbuf

>From: Michael Qiu [mailto:qdy220091...@gmail.com] >Sent: Monday, June 19, 2017 6:29 AM >To: d...@openvswitch.org >Cc: Kavanagh, Mark B ; b...@ovn.org; >db...@vmware.com; Michael Qiu >Subject: [PATCH 4/5] lib/dp-packet: copy multi-segments data from

Re: [ovs-dev] [PATCH] datapath: Introduce reference counter in datapath tunnel port

Hi, I have two questions regarding the tunnel handling. These questions are general questions as the behavior is existing in the vanilla code, not introduced by this patch. These are potential areas in the code where change can be needed. I would like to reconfigure the code in separate

[ovs-dev] [PATCH] datapath: Introduce reference counter in datapath tunnel port

There is a problem with deleting one out of more OF tunnel ports with the same port type. In OVS all OF tunnel ports sharing the same tunnel type and dst_port setting are mapped to a common tunnel port instance in the datapath. Available tunnel types are GRE, VXLAN, GENEVE and STT. dst_port is

Re: [ovs-dev] [RFC PATCH v1 1/3] vswitch.xml: Detail ipsec user interface.

> > Regards > _Sugesh > > > > -Original Message- > > From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev- > > boun...@openvswitch.org] On Behalf Of Ian Stokes > > Sent: Thursday, May 18, 2017 10:16 AM > > To: d...@openvswitch.org > > Subject: [ovs-dev] [RFC PATCH v1 1/3] vswitch.xml:

Re: [ovs-dev] [RFC PATCH v1 0/3] RFC for Userspace IPsec Interface

> Hi Ian, > Thank you for working on this! > Good initiative and its definitely very useful. > Few high level comments as below. > Have added some more comments in the following patches as well. > > > Regards > _Sugesh > > > > -Original Message- > > From:

[ovs-dev] [PATCH] ofproto-dpif-ipfix: add Exporting Process Reliability Statistics

This patch adds support for The Exporting Process Reliability Statistics Option Template, which is defined in RFC 5101, Section 4.3. For now, this Option Template includes only exporting process ID scope field and a single counter - notSentPacketTotalCount. Currently, Data Records that contain

[ovs-dev] [PATCH v3 6/6] ovn: l3ha, gratuitous ARP for HA router

From: Venkata Anil Kommaddi This patch extends gratuitous ARP support for NAT addresses so that it applies to centralized NAT rules on a HA router. Gratuitous ARP packets for centralized NAT rules on a HA router are only generated on the active gateway chassis.

[ovs-dev] [PATCH v3 5/6] ovn: l3ha, make is_chassis_active aware of gateway_chassis

is_chassis_active now is only true for redirect-chassis ports in the case of the specific lport being active on the local chassis. This will naturally make the ARP responder / redirection openflow rules automatically inserted/removed when a router goes active/backup. Signed-off-by: Miguel Angel

[ovs-dev] [PATCH v3 4/6] ovn: l3ha, enable bfd between tunnel endpoints

From: Venkata Anil This patch enables bfd protocol between gateways and transport nodes, all gateway nodes with any HA chassisredirect port will enable BFD to all tunnel endpoints, while transport nodes will enable BFD to all tunnel endpoints hosting an HA gateway

[ovs-dev] [PATCH v3 3/6] ovn: l3ha, handling of multiple gateway chassis

This patch handles multiple gateway_chassis with in chassisredirect ports, any gateway with a chassis redirect port will implement the rules to de-encapsulate incomming packets for such port. And hosts targetting a remote chassisredirect port will setup a bundle(active_backup, ..) action to each

[ovs-dev] [PATCH v3 2/6] ovn: l3ha, ovn-northd gateway chassis propagation

The redirect-chassis option of logical router ports is now translated to Gateway_Chassis entries for backwards compatibility. Gateway_Chassis entries in nbdb are copied over to sbdb and linked them to the Chassis entry. Signed-off-by: Miguel Angel Ajo Signed-off-by: Anil

[ovs-dev] [PATCH v3 1/6] ovn: l3ha, NBDB and SBDB changes and documentation

This commit introduces the north and south db changes necessary for the l3ha router implementation. It defines a new Table in both NBDB and SBDB. The Gateway_Chassis table is created, with a tiny difference between NBDB and SBDB, NBDB references the chassis via it's name (chassis_name) and SBDB

[ovs-dev] Pricelist 05-07-2017

Deze email nieuwsbrief werd in grafisch HTML formaat verzonden. Als u deze tekstversie ziet, verkiest uw email programma "gewone tekst" emails. U kan de originele nieuwsbrief online bekijken:

[ovs-dev] [PATCH 06/11] ovs/dp-cls: remove data-path classifier rule

PMD send a flow that would be removed by HW classifier Signed-off-by: Shachar Beiser Conflicts: lib/hw-pipeline.c --- lib/dpif-netdev.c | 9 lib/hw-pipeline.c | 149 ++ lib/hw-pipeline.h | 14 + 3

[ovs-dev] [PATCH 02/11] ovs/dp-cls: moving structures to dpif-netdev.h

hw-pipeline.c , dpif-netdev.c share structures & functions. In order to reuse them ,they are moved to dpif-netdev.h Signed-off-by: Shachar Beiser --- lib/dpif-netdev.c | 175 +--- lib/dpif-netdev.h | 298

[ovs-dev] [PATCH 08/11] ovs/dp-cls: tag lookup and processing

PMD fetch the flow according to the flow tag . Signed-off-by: Shachar Beiser Conflicts: lib/dpif-netdev.c lib/hw-pipeline.c --- lib/dpif-netdev.c | 58 +-- lib/hw-pipeline.c | 115

[ovs-dev] [PATCH 04/11] ovs/dp-cls: initializing HW pipeline

The HW pipeline is made of 3 entites: dp-cls thread, a pool of flow tags and a message queue between the pmd context and dp-cls offload thread. This patch initializes those 3 entities Signed-off-by: Shachar Beiser --- lib/automake.mk | 1 + lib/dpif-netdev.c | 1 +

[ovs-dev] [PATCH 10/11] ovs/dp-cls: removing flow from HW in the dp-cls offload thread

offload dp-cls thread destroy flow in the HW classifier Signed-off-by: Shachar Beiser Conflicts: lib/hw-pipeline.c --- lib/hw-pipeline.c | 74 +++ 1 file changed, 74 insertions(+) diff --git

[ovs-dev] [PATCH 05/11] ovs/dp-cls: free HW pipeline

The HW pipeline is made of 3 entites: dp-cls thread, a pool of flow tags and a message queue between the pmd context and dp-cls offload thread. This patch frees those 3 entities. Signed-off-by: Shachar Beiser --- lib/dpif-netdev.c | 7 +-- lib/hw-pipeline.c | 60

[ovs-dev] [PATCH 07/11] ovs/dp-cls: inserting data-path classifier rule

PMD send a flow that would be inserted by HW classifier Signed-off-by: Shachar Beiser Conflicts: lib/dpif-netdev.c --- lib/dpif-netdev.c | 30 lib/hw-pipeline.c | 101 ++ 2 files changed, 124

[ovs-dev] [PATCH 01/11] ovs/dp-cls: fetching the mark id from hw

The HW set the mark id that represents matching rule. The hw-pipeline reads the mark id from fdir.hi Signed-off-by: Shachar Beiser --- lib/automake.mk | 3 ++- lib/hw-pipeline.h | 31 +++ lib/netdev-bsd.c | 1 +

[ovs-dev] [PATCH 09/11] ovs/dp-cls: flow tag read

PMD thread read the flow tag from the rte_mbuf . Signed-off-by: Shachar Beiser --- lib/dpif-netdev.c | 30 +- lib/hw-pipeline.c | 11 +++ 2 files changed, 28 insertions(+), 13 deletions(-) diff --git a/lib/dpif-netdev.c

[ovs-dev] [PATCH 03/11] ovs/dp-cls: saving rx queue identifier

The receive queue id is saved in order to direct the data-path classifier traffic towards it. Signed-off-by: Shachar Beiser --- lib/dp-packet.h | 1 + lib/netdev.c| 2 ++ 2 files changed, 3 insertions(+) diff --git a/lib/dp-packet.h b/lib/dp-packet.h index

[ovs-dev] [PATCH 00/11] Data Path Classifier Offloading

Request for comments Subject: OVS data-path classifier offload Versions: OVS master. DPDK 17.05.1. Purpose & Scope This RFC describes the flows’ hardware offloading over DPDK . The motivation of hardware flows’ offloading is accelerating the OVS DPDK data plane.The classification is done by

[ovs-dev] [PATCH] netdev: should check ifa_name too

the ifa_name is null and ovs-vswitchd crashed. but i don't know how this happen. the stack is : hread 1 (LWP 2275): (gdb) x/16i $pc => 0x7f25f89d235c: movdqu (%rdi),%xmm1 0x7f25f89d2360: movdqu (%rsi),%xmm2 0x7f25f89d2364: pxor %xmm0,%xmm0 0x7f25f89d2368: pcmpeqb

Re: [ovs-dev] GTP support to OVS upstream

I remember OpenFlow 1.6 spec (not finalized) proposes to use OpenFlow actions to do GTP-u encapsulation and decapsulation, current OvS tunnel implementation can't support the third kind of use case (don't encap & decap, just parse, match and forward), MEC (ETSI Molibe/Multi-access Edge

Re: [ovs-dev] [PATCH] acinclude: netdev-dpdk: use pkg-config of libdpdk

On Wed, 2017-07-05 at 11:19 +0200, Christian Ehrhardt wrote: > If available use dpdk pkg-config info of libdpdk to set the right > include paths. > That for example, allows packagers to provide non default include > paths in a common way (pkg-config). > > Signed-off-by: Christian Ehrhardt

Re: [ovs-dev] [PATCH ovs] dpif: fix warn msg when failed to open netdev

On 4 July 2017 at 11:04, Marcelo Ricardo Leitner wrote: > Currently it is using the datapath name/type but what has actually > failed was the netdev. > > Fix it by using netdev name/type instead and also log why it failed. > > Signed-off-by: Marcelo Ricardo Leitner

Re: [ovs-dev] [PATCH] odp: Fix crash in parse_8021q_onward().

On 3 July 2017 at 13:17, Eric Garver wrote: > On Fri, Jun 16, 2017 at 04:51:04AM -0700, nickcooper-zhangtonghao wrote: >> When we use the 'ovs-appctl ofproto/trace' to send packets, >> which include the 'vlan' field, but exclude the 'encap', >> the ovs-vswitchd will crash. We should

[ovs-dev] GTP support to OVS upstream

Dear Developers, I would like to start a discussion to actually get native GTP tunneling support in Openvswitch keeping it still compatible with the current OpenFlow standard. https://en.wikipedia.org/wiki/GPRS_Tunnelling_Protocol I the past there has been a patch for OVS to actually have GTP

Re: [ovs-dev] [PATCH v3] netdev-dpdk: use rte_eth_dev_set_mtu

>From: Chandran, Sugesh >Sent: Monday, July 3, 2017 4:36 PM >To: Kavanagh, Mark B ; ovs-dev@openvswitch.org; >Varghese, Vipin ; acon...@redhat.com >Subject: RE: [ovs-dev] [PATCH v3] netdev-dpdk: use rte_eth_dev_set_mtu > >Hi Mark, > >Regards

[ovs-dev] [PATCH] acinclude: netdev-dpdk: use pkg-config of libdpdk

If available use dpdk pkg-config info of libdpdk to set the right include paths. That for example, allows packagers to provide non default include paths in a common way (pkg-config). Signed-off-by: Christian Ehrhardt Suggested-by: Luca Boccassi

[ovs-dev] [patch_v3 7/8] System tests: Add a few passive ftp tests.

In order to have full coverage of ALGs for the userspace datapath, it is necessary to add 3 tests covering passive ftp. Before adding these new tests, there was only one part of one test that covered passive ftp. The new tests cover basic V6 passive ftp, V4 passive ftp with NAT and sequence skew

[ovs-dev] [patch_v3 8/8] NEWS: Announce userspace datapath ALG support.

Signed-off-by: Darrell Ball --- NEWS | 1 + 1 file changed, 1 insertion(+) diff --git a/NEWS b/NEWS index 0f2604f..fc5c8f4 100644 --- a/NEWS +++ b/NEWS @@ -29,6 +29,7 @@ Post-v2.7.0 abbreviated to 4 hex digits. - Userspace Datapath: * Added NAT support for

[ovs-dev] [patch_v3 4/8] Userspace Datapath: Add ALG infra and FTP.

ALG infra and FTP (both V4 and V6) support is added to the userspace datapath. Also, NAT support is included. Signed-off-by: Darrell Ball --- lib/conntrack-private.h | 20 +- lib/conntrack.c | 1027 +++ lib/conntrack.h

[ovs-dev] [patch_v3 6/8] System tests: Enable ALGs for userspace.

Signed-off-by: Darrell Ball --- tests/system-userspace-macros.at | 7 ++- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/tests/system-userspace-macros.at b/tests/system-userspace-macros.at index 6e3d468..417aa4f 100644 --- a/tests/system-userspace-macros.at

[ovs-dev] [patch_v3 5/8] Userspace Datapath: Add TFTP support.

Both ipv4 and ipv6 are supported. Also, NAT support is included. Signed-off-by: Darrell Ball --- include/sparse/netinet/in.h | 1 + lib/conntrack.c | 45 - 2 files changed, 45 insertions(+), 1 deletion(-) diff --git

[ovs-dev] [patch_v3 1/8] Userspace Datapath: Change 'related' to 'icmp_related'.

The conntrack context flag 'related' is changed to 'icmp_related' to disambiguate usage w.r.t. ALGs which are added in a subsequent patch. Signed-off-by: Darrell Ball --- lib/conntrack.c | 12 ++-- 1 file changed, 6 insertions(+), 6 deletions(-) diff --git

[ovs-dev] [patch_v3 0/8] Userspace Datapath: Add ALG support.

ALG infra is added with support for FTP and TFTP. Both V4 and V6 are supported. Also, NAT is supported. Three passive ftp system tests are added to complete testing coverage of ftp for the userspace datapath, as the existing coverage of passive ftp was limited to one part of one test for V4