Re: [ovs-dev] [PATCH v9 0/7] OVS-DPDK flow offload with rte_flow

> Hi, > > Here is a joint work from Mellanox and Napatech, to enable the flow hw > offload with the DPDK generic flow interface (rte_flow). Hi folks, I feel Mellanox/Netronome have reached the point where HWOL can be introduced to OVS DPDK pending performance review. This has not been an easy

Re: [ovs-dev] [PATCH V2] datapath: Prevent panic

On 4/19/2018 4:18 PM, Flavio Leitner wrote: On Tue, Apr 17, 2018 at 12:34:08PM -0700, Greg Rose wrote: On RHEL 7.x kernels we observe a panic induced by a paging error when the timer kicks off a job that subsequently accesses memory that belonged to the openvswitch kernel module but was since

Re: [ovs-dev] [PATCH v2] ovs-ofctl: Fixed the "snoop" command of ovs-ofctl

Sure, I will update the comment in the code to explain the effect of using: "use_names = 0". Also, would change the comment in the commit back to the original message. Thanks, On Wed, Apr 18, 2018 at 4:29 PM, Ben Pfaff wrote: > On Tue, Apr 17, 2018 at 06:40:46PM -0700, Ashish

Re: [ovs-dev] [PATCH V2] datapath: Prevent panic

On Tue, Apr 17, 2018 at 12:34:08PM -0700, Greg Rose wrote: > On RHEL 7.x kernels we observe a panic induced by a paging error > when the timer kicks off a job that subsequently accesses memory > that belonged to the openvswitch kernel module but was since > unloaded - thus the paging error. > >

Re: [ovs-dev] OVS upgrade

On Thu, Apr 19, 2018 at 03:32:28PM -0400, Amir Mohamad wrote: > I have openstack installed on one host. I have one controller (VM) and > one Compute node (VM). > I want to install a new version that supports OVN SFC by cloning this repo > https://github.com/doonhammer/ovs.git Why that one?

[ovs-dev] OVS upgrade

Hi, I have openstack installed on one host. I have one controller (VM) and one Compute node (VM). I want to install a new version that supports OVN SFC by cloning this repo https://github.com/doonhammer/ovs.git In the old OVS installation: 1. ovsdb-server and ovs-vswitchd are in /usr/sbin 2.

Re: [ovs-dev] [PATCH v2 0/4] Fix URLS and and associated markups.

> On Thu, Apr 19, 2018 at 03:17:08PM +0100, Ian Stokes wrote: > > Running make check-docs has caught a number of URLs that are no longer > > active. This patch series updates the required URLs along with some > > mark-up changes in the docs in relation to said URLs. > > > > I originally looked at

Re: [ovs-dev] [PATCH v2 1/1] netdev-dpdk: fix RX jumbo for NICs not supporting scatter

On 04/19/2018 03:32 PM, Pablo Cascón wrote: > On 18/04/18 18:35, Kevin Traynor wrote: >> On 04/18/2018 03:41 PM, Pablo Cascón wrote: >>> On 13/04/18 19:45, Kevin Traynor wrote: On 04/13/2018 04:20 PM, Stokes, Ian wrote: >> Currently to RX jumbo packets fails for NICs not supporting

[ovs-dev] Could be a bug: extremely long latency for ovn-controller to get SB update in scale test

While working on ovn-controller incremental processing, I was testing the end-to-end latency and comparing with master. I found the latency with master branch is extremely long while changing the way of testing makes it much shorter. The master branch was at commit 593e93e5f (ovsdb clustering was

Re: [ovs-dev] [PATCH] ofproto: Allow bundle idle timeout to be configured.

On Fri, Apr 13, 2018 at 10:48:09AM -0700, Ben Pfaff wrote: > On Fri, Apr 13, 2018 at 01:45:30PM -0300, Flavio Leitner wrote: > > In some cases 10 seconds might be too much time and in > > other cases it might be too little. > > > > The OpenFlow spec mandates that it should wait at least one > >

[ovs-dev] [PATCH v2] ofproto: Allow bundle idle timeout to be configured.

In some cases 10 seconds might be too much time and in other cases it might be too little. The OpenFlow spec mandates that it should wait at least one second, so enforce that as the minimum acceptable value. Signed-off-by: Flavio Leitner --- ofproto/connmgr.c | 24

Re: [ovs-dev] [PATCH] ofproto-dpif-upcall: Only call ovsrcu_postpone() on active actions

On Thu, Apr 19, 2018 at 01:24:06PM +0200, Eelco Chaudron wrote: > Currently, ovsrcu_postpone() is called even with a NULL argument, > i.e. when there is no data to be freed. This is causing additional > overhead because work is scheduled for the urcu thread. This change > avoids adding the

Re: [ovs-dev] [PATCH v2] python: Tighten the check if we need encoding

On Thu, Apr 19, 2018 at 11:33:21AM +0200, Jakub Sitnicki wrote: > Check if we are dealing with a Unicode string that needs > encoding for both Python 2 & 3. > > Also, do the encoding the same way for Python 2 & 3 and avoid using > negation to make the code simpler. > > v1 -> v2: Rewrite the

Re: [ovs-dev] [PATCH v2 0/4] Fix URLS and and associated markups.

On Thu, Apr 19, 2018 at 03:17:08PM +0100, Ian Stokes wrote: > Running make check-docs has caught a number of URLs that are no longer > active. This patch series updates the required URLs along with some mark-up > changes in the docs in relation to said URLs. > > I originally looked at doing this

[ovs-dev] [PATCH v3] rhel: user/group openvswitch does not exist

From: Alan Pevec Default ownership[1] for config files is failing on an empty system: Running scriptlet: openvswitch-2.9.0-3.fc28.x86_64 warning: user openvswitch does not exist - using root warning: group openvswitch does not exist - using root ... Required user/group

Re: [ovs-dev] [PATCH v3 2/9] doc: Add "PMD" topic document

On 04/19/2018 01:57 PM, Stephen Finucane wrote: > This continues the breakup of the huge DPDK "howto" into smaller > components. There are a couple of related changes included, such as > using "Rx queue" instead of "rxq" and noting how Tx queues cannot be > configured. > > Signed-off-by: Stephen

Re: [ovs-dev] [PATCH v2 1/1] netdev-dpdk: fix RX jumbo for NICs not supporting scatter

On 18/04/18 18:35, Kevin Traynor wrote: On 04/18/2018 03:41 PM, Pablo Cascón wrote: On 13/04/18 19:45, Kevin Traynor wrote: On 04/13/2018 04:20 PM, Stokes, Ian wrote: Currently to RX jumbo packets fails for NICs not supporting scatter. Scatter is not strictly needed for jumbo support on RX.

[ovs-dev] [PATCH v2 4/4] docs: Fix urls in index.rst.

This patch prepends 'www' to openvswitch urls in index.rst. Without this make check-docs fails when verifying url liveness. Also remove url referencing ovsdb-server(5) as these are no longer accessible. Cc: Stephen Finucane Fixes: 4f6ec357c ("doc: Populate 'ref' section")

[ovs-dev] [PATCH v2 3/4] docs: Fix sphinx urls.

Update dead url links for sphinx documentation to avoid make check-docs failing. Cc: Stephen Finucane Fixes: 26ea2d409 ("docs: Add writing guide") Fixes: 73c76b447 ("doc: Add info on building documentation") Signed-off-by: Ian Stokes Acked-by: Stephen

[ovs-dev] [PATCH v2 2/4] docs: Fix style guide url in DocumentationStyle.rst.

The link used for IBM Style Guide is no longer valid. As there is no longer a valid link via redbooks remove the url to avoid make check-docs failing. Cc: Stephen Finucane Fixes: 26ea2d409 ("docs: Add writing guide") Signed-off-by: Ian Stokes Acked-by:

[ovs-dev] [PATCH v2 1/4] docs: Fix sflow documentation url and markup.

The link url link for the blog in sflow documentation causes make check-docs to fail with a broken link warning. Fix this by correcting the url address. Also use correct markup for note regarding the configuration of sflow. CC: Stephen Finucane Fixes: 198c5d3d0 ("doc: Add

[ovs-dev] [PATCH v2 0/4] Fix URLS and and associated markups.

Running make check-docs has caught a number of URLs that are no longer active. This patch series updates the required URLs along with some mark-up changes in the docs in relation to said URLs. I originally looked at doing this in a single patch however as the URLs are spread throughout the

[ovs-dev] [PATCH v3 6/9] doc: Add "bridge" topic document

This details configuration steps that apply to the entire bridge, rather than individual ports. Signed-off-by: Stephen Finucane --- v3: - Add new files to automake.mk v2: - Cross-reference this document from all interface documents --- Documentation/automake.mk

[ovs-dev] [PATCH v3 2/9] doc: Add "PMD" topic document

This continues the breakup of the huge DPDK "howto" into smaller components. There are a couple of related changes included, such as using "Rx queue" instead of "rxq" and noting how Tx queues cannot be configured. Signed-off-by: Stephen Finucane --- v3: - Add new files to

[ovs-dev] [PATCH v3 4/9] doc: Add "vdev" topic document

These are separate things from physical, ring and vhost-user interfaces and deserve their own documents. A couple of small typos are fixed along the way. Signed-off-by: Stephen Finucane --- v3: - Add new files to automake.mk - Add 'versionchanged' directive to call out when

[ovs-dev] [PATCH v3 8/9] doc: Add "jumbo frames" topic document

We include references from the physical and vhost-user interface guides. Signed-off-by: Stephen Finucane --- v3: - Add new files to automake.mk - Add 'versionchanged' directive to call out when this was added v2: - Don't split the document into multiple docs ---

[ovs-dev] [PATCH v3 3/9] doc: Move additional sections to "physical ports" doc

The "hotplugging", "flow control", and "Rx checksum offload" sections only apply to 'dpdk' ports and are too detailed to include in a high-level howto. Move them, reworking some aspects of this in the process. Signed-off-by: Stephen Finucane --- v2: - Split out vdev into its

[ovs-dev] [PATCH v3 1/9] doc: Add an overview of the 'dpdk' port

These ports are used to allow ingress/egress from the host and are therefore _reasonably_ important. However, there is no clear overview of what these ports actually are or why things are done the way they are. Start closing this gap by providing a standalone example of using these ports along

[ovs-dev] [PATCH v3 5/9] doc: Move "QoS" guide to its own document

Again, this stuff is too detailed for a high-level howto. Signed-off-by: Stephen Finucane --- v3: - Add new files to automake.mk - Add 'versionchanged' directive to call out when this was added v2: - Keep flow control section in 'phy' document --- Documentation/automake.mk

[ovs-dev] [PATCH v3 0/9] Split up the DPDK how-to

The DPDK howto has slowly morphed into a catch all for everything DPDK, which goes against the original design goal for 'howto' documents [*]. This series attempts to return some sanity to the universe by splitting this document into many more 'topic' documents. Along the way, we add a lot of

Re: [ovs-dev] [PATCH v1 3/4] docs: Fix sphinx urls.

> On Thu, 2018-04-19 at 11:49 +0100, Ian Stokes wrote: > > Update dead url links for sphinx documentation to avoid make > > check-docs failing. > > > > Cc: Stephen Finucane > > Fixes: 26ea2d409 ("docs: Add writing guide") > > Fixes: 73c76b447 ("doc: Add info on building

Re: [ovs-dev] [PATCH v2 6/9] doc: Add "bridge" topic document

On Wed, 2018-04-18 at 15:33 +, Stokes, Ian wrote: > > This details configuration steps that apply to the entire bridge, rather > > than individual ports. > > > > Signed-off-by: Stephen Finucane > > --- > > v2: > > - Cross-reference this document from all interface

Re: [ovs-dev] [PATCH v1 1/4] Docs: Fix sflow documentation url and markup.

> -Original Message- > From: Stephen Finucane [mailto:step...@that.guru] > Sent: Thursday, April 19, 2018 1:25 PM > To: Stokes, Ian ; d...@openvswitch.org > Subject: Re: [PATCH v1 1/4] Docs: Fix sflow documentation url and markup. > > On Thu, 2018-04-19 at 11:49

Re: [ovs-dev] [PATCH v1 2/4] docs: Fix style guide url in DocumentationStyle.rst.

> On Thu, 2018-04-19 at 11:49 +0100, Ian Stokes wrote: > > The link used for IBM Style Guide is no longer valid. As there is no > > longer a valid link via redbooks remove the url to avoid make > > check-docs failing. > > > > Cc: Stephen Finucane > > Fixes: 26ea2d409 ("docs:

Re: [ovs-dev] [PATCH v2 2/9] doc: Add "PMD" topic document

On Wed, 2018-04-18 at 15:31 +, Stokes, Ian wrote: > > This continues the breakup of the huge DPDK "howto" into smaller > > components. There are a couple of related changes included, such as using > > "Rx queue" instead of "rxq" and noting how Tx queues cannot be configured. > > > >

Re: [ovs-dev] [PATCH v2 4/9] doc: Add "vdev" topic document

On Wed, 2018-04-18 at 15:32 +, Stokes, Ian wrote: > > These are separate things from physical, ring and vhost-user interfaces > > and deserve their own documents. A couple of small typos are fixed along > > the way. > > > > Signed-off-by: Stephen Finucane > > --- > >

Re: [ovs-dev] [PATCH v2 1/9] doc: Add an overview of the 'dpdk' port

On Wed, 2018-04-18 at 15:30 +, Stokes, Ian wrote: > > These ports are used to allow ingress/egress from the host and are > > therefore _reasonably_ important. However, there is no clear overview of > > what these ports actually are or why things are done the way they are. > > Start closing

Re: [ovs-dev] [PATCH v1 3/4] docs: Fix sphinx urls.

On Thu, 2018-04-19 at 11:49 +0100, Ian Stokes wrote: > Update dead url links for sphinx documentation to avoid > make check-docs failing. > > Cc: Stephen Finucane > Fixes: 26ea2d409 ("docs: Add writing guide") > Fixes: 73c76b447 ("doc: Add info on building documentation") >

Re: [ovs-dev] [PATCH v1 2/4] docs: Fix style guide url in DocumentationStyle.rst.

On Thu, 2018-04-19 at 11:49 +0100, Ian Stokes wrote: > The link used for IBM Style Guide is no longer valid. As there is no > longer a valid link via redbooks remove the url to avoid make > check-docs failing. > > Cc: Stephen Finucane > Fixes: 26ea2d409 ("docs: Add writing

Re: [ovs-dev] [PATCH v1 4/4] docs: Fix urls in index.rst.

On Thu, 2018-04-19 at 11:49 +0100, Ian Stokes wrote: > This patch prepends 'www.' to openvswitch urls in index.rst. Without this > make check-docs fails when verifying url liveness. Also remove url > referencing ovsdb-server(5) as these are no longer accessible. > > Cc: Stephen Finucane

Re: [ovs-dev] [PATCH v1 1/4] Docs: Fix sflow documentation url and markup.

On Thu, 2018-04-19 at 11:49 +0100, Ian Stokes wrote: > The link url link for the blog in sflow documentation causes make > check-docs to fail with a broken link warning. Fix this by correcting > the url address. Also use correct markup for note regarding the > configuration of sflow. > > CC:

[ovs-dev] [PATCH] ofproto-dpif-upcall: Only call ovsrcu_postpone() on active actions

Currently, ovsrcu_postpone() is called even with a NULL argument, i.e. when there is no data to be freed. This is causing additional overhead because work is scheduled for the urcu thread. This change avoids adding the postpone callback if no work needs to be done. This especially helps for the

[ovs-dev] [PATCH v1 4/4] docs: Fix urls in index.rst.

This patch prepends 'www.' to openvswitch urls in index.rst. Without this make check-docs fails when verifying url liveness. Also remove url referencing ovsdb-server(5) as these are no longer accessible. Cc: Stephen Finucane Fixes: 4f6ec357c ("doc: Populate 'ref' section")

[ovs-dev] [PATCH v1 3/4] docs: Fix sphinx urls.

Update dead url links for sphinx documentation to avoid make check-docs failing. Cc: Stephen Finucane Fixes: 26ea2d409 ("docs: Add writing guide") Fixes: 73c76b447 ("doc: Add info on building documentation") Signed-off-by: Ian Stokes ---

[ovs-dev] [PATCH v1 2/4] docs: Fix style guide url in DocumentationStyle.rst.

The link used for IBM Style Guide is no longer valid. As there is no longer a valid link via redbooks remove the url to avoid make check-docs failing. Cc: Stephen Finucane Fixes: 26ea2d409 ("docs: Add writing guide") Signed-off-by: Ian Stokes ---

[ovs-dev] [PATCH v1 1/4] Docs: Fix sflow documentation url and markup.

The link url link for the blog in sflow documentation causes make check-docs to fail with a broken link warning. Fix this by correcting the url address. Also use correct markup for note regarding the configuration of sflow. CC: Stephen Finucane Fixes: 198c5d3d0 ("doc: Add

[ovs-dev] [PATCH v1 0/4] Fix URLS and and associated markups.

Running make check-docs has caught a number of URLs that are no longer active. This patch series updates the required URLs along with some mark-up changes in the docs in relation to said URLs. I originally looked at doing this in a single patch however as the URLs are spread throughout the

Re: [ovs-dev] [PATCH v2] rhel: user/group openvswitch does not exist

On Wed, Apr 18, 2018 at 5:54 PM, Aaron Conole wrote: > I can certainly re-add it. The rationale section calls adding the 'exit > 0' "...suboptimal but has less potential for system wide breakage..." > > I can go either way. Maybe Alan or Timothy have opinions? I'd stick to

Re: [ovs-dev] [PATCH 2/2] dpdk docs: Drop file share in libvirt config.

> On Thu, 2018-04-12 at 08:24 +0100, Lam, Tiago wrote: > > On 11/04/2018 15:03, Stephen Finucane wrote: > > > On Wed, 2018-04-11 at 09:54 -0400, Aaron Conole wrote: > > > > Tiago Lam writes: > > > > > > > > > When explaining on how to add vhost-user ports to a guest, using >

Re: [ovs-dev] [PATCH 1/2] dpdk docs: Drop qemu-kvm for qemu-system-x86_64.

> Tiago Lam writes: > > > When explaining on how to add vhost-user ports to a guest, using > > libvirt, point to the qemu-system-x86_64 binary by default, instead of > > using qemu-kvm. The latter has been made obsolete and dropped from a > > number of distributions