Re: [ovs-dev] [PATCH ovn 1/2 v2] Add ovn-ctl to assist with OVN daemon lifecycle.

On Mon, May 04, 2015 at 05:40:45PM -0400, Russell Bryant wrote: > This patch introduces ovn-ctl, which is similar to ovs-ctl. I opted > for a new script as everything in OVN so far is nicely isolated, so a > new script seemed to make the most sense. > > ovn-ctl supports a few different scenarios.

Re: [ovs-dev] [PATCH ovn 2/2 v2] fedora.spec: Create openvswitch-ovn package.

On Mon, May 04, 2015 at 05:40:46PM -0400, Russell Bryant wrote: > This patch creates a new subpackage for OVN, openvswitch-ovn. It also > installs systemd unit files for ovn-controller and ovn-northd. > > If you want to run ovn-controller: > > # systemctl start ovncontroller > > If you want

[ovs-dev] [PATCH v7] 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 ('

[ovs-dev] [PATCH v3 4/4] netdev-dpdk: Adapt the requested number of tx and rx queues.

This commit changes the semantics of 'netdev_set_multiq()' to allow OVS DPDK to run on device with limited multi queue support. * If a netdev doesn't have the requested number of rxqs it can simply inform the datapath without failing. * If a netdev doesn't have the requested number of txqs it sh

[ovs-dev] [PATCH v3 3/4] netdev-dpdk: Use specific spinlock for stats.

Right now ethernet and ring devices use a mutex, while vhost devices use a mutex or a spinlock to protect statistics. This commit introduces a single spinlock that's always used for stats updates. Signed-off-by: Daniele Di Proietto --- lib/netdev-dpdk.c | 29 + 1 fil

[ovs-dev] [PATCH v3 1/4] ovs-numa: Change 'core_id' to unsigned.

DPDK lcore_id is unsigned. We need to support big values like LCORE_ID_ANY (=UINT32_MAX). Therefore I am changing the type everywhere in OVS. Signed-off-by: Daniele Di Proietto --- lib/dpif-netdev.c | 27 +++ lib/dpif.c| 6 +++--- lib/dp

[ovs-dev] [PATCH v3 0/4] DPDK: simplify configuration

This series improves OVS configuration with DPDK in two ways: * netdev-dpdk is patched to work with smaller NICs (without lots of transmission queues). * DPDK non-EAL threads are used: with this change we can avoid reserving core 0 for non PMD operations v2 -> v3: * I've dropped the patch t

[ovs-dev] [PATCH v3 2/4] netdev-dpdk: Properly support non pmd threads.

We used to reserve DPDK lcore 0 for non pmd operations, making it difficult to use core 0 for packet processing. DPDK 2.0 properly support non EAL threads with lcore LCORE_ID_ANY. Using non EAL threads for non pmd threads, we do not need to reserve any core for non pmd operations Signed-off-by: D

Re: [ovs-dev] [PATCH v7] netdev-dpdk: add dpdk vhost-user ports

On 5/22/15 11:40 AM, Ciara Loftus wrote: This patch adds support for a new port type to the userspace datapath called dpdkvhostuser. It looks good to me. It is much needed too! A new dpdkvhostuser port will create a unix domain socket which when provided to QEMU is used to facilitate communica

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

I was running an old version, getting it from the upstream repo helped, thanks! I manage to compile everything without warnings if I apply the following incremental ---8<--- diff --git a/include/sparse/automake.mk b/include/sparse/automake.mk index c80c4c2..0456ee6 100644 --- a/include/sparse/au

[ovs-dev] [PATCH] dpctl: Don't print UFID if not present.

With verbose dpctl, if userspace runs against an older kernel, every entry will have "ufid:" at the beginning. This is unnecessary and introduces an additional format for scripts to parse. Drop it. Signed-off-by: Joe Stringer --- lib/dpctl.c | 10 +++--- 1 file changed, 3 insertions(+), 7 de

Re: [ovs-dev] [PATCH v3 1/4] ovs-numa: Change 'core_id' to unsigned.

Acked-by: Ethan Jackson On Fri, May 22, 2015 at 9:14 AM, Daniele Di Proietto wrote: > DPDK lcore_id is unsigned. We need to support big values like > LCORE_ID_ANY (=UINT32_MAX). Therefore I am changing the type everywhere > in OVS. > > Signed-off-by: Daniele Di Proietto > --- > lib/dpif-net

Re: [ovs-dev] [PATCH v3 2/4] netdev-dpdk: Properly support non pmd threads.

Acked-by: Ethan Jackson On Fri, May 22, 2015 at 9:14 AM, Daniele Di Proietto wrote: > We used to reserve DPDK lcore 0 for non pmd operations, making it > difficult to use core 0 for packet processing. > DPDK 2.0 properly support non EAL threads with lcore LCORE_ID_ANY. > > Using non EAL threads

Re: [ovs-dev] [PATCH v3 3/4] netdev-dpdk: Use specific spinlock for stats.

Acked-by: Ethan Jackson On Fri, May 22, 2015 at 9:14 AM, Daniele Di Proietto wrote: > Right now ethernet and ring devices use a mutex, while vhost devices use > a mutex or a spinlock to protect statistics. This commit introduces a > single spinlock that's always used for stats updates. > > Sig

Re: [ovs-dev] [PATCH v5 1/4] datapath-windows: Support for custom VXLAN tunnel port

> On May 21, 2015, at 12:54 PM, Sorin Vinturis > wrote: > > Nithin, > > I thought I have addressed your comment. My understanding was that was an > issue in OvsDeleteVportCmdHandler() after the call to > OvsRemoveAndDeleteVport(). The latter function might deallocate the vport, > which woul

Re: [ovs-dev] [PATCH v3 4/4] netdev-dpdk: Adapt the requested number of tx and rx queues.

In netdev_dpdk_send__() I folded in an OVS_UNLIKELY around the txq_needs_locking check as I'd like to optimize for the standard case where we don't. Otherwise looks good to me, I'll merge this series shortly. Acked-by: Ethan Jackson On Fri, May 22, 2015 at 9:14 AM, Daniele Di Proietto wrote:

Re: [ovs-dev] [PATCH] dpctl: Don't print UFID if not present.

On Fri, May 22, 2015 at 10:32:53AM -0700, Joe Stringer wrote: > With verbose dpctl, if userspace runs against an older kernel, every > entry will have "ufid:" at the beginning. This is unnecessary and > introduces an additional format for scripts to parse. Drop it. > > Signed-off-by: Joe Stringer

Re: [ovs-dev] [PATCH] ovn-northd: Add appctl support.

On Thu, May 21, 2015 at 03:20:48PM -0700, Alex Wang wrote: > This commit adds appctl support to ovn-northd and the 'exit' command. > The 'exit' command can be used in autotest for stopping the ovn-northd. > > Signed-off-by: Alex Wang Acked-by: Ben Pfaff _

Re: [ovs-dev] [PATCH V2] ovn: Change type of 'chassis' column in 'Binding' table.

Fine with me. On Thu, May 21, 2015 at 10:02:12AM -0700, Alex Wang wrote: > How about adding an inline function like this?~ > > On Thu, May 21, 2015 at 10:04 AM, Alex Wang wrote: > > > This commit changes the type of 'chassis' column in 'Binding' table > > from string to weak reference of 'Chass

Re: [ovs-dev] [PATCH] datapath: Support masked set actions.

> On May 21, 2015, at 9:23 PM, Jesse Gross wrote: > > On Thu, May 21, 2015 at 4:39 PM, Jarno Rajahalme > wrote: >> OVS kernel module support for masked set actions in already upstream >> in Linux. This patch adds the same for the OVS tree kernel module. >> >> The

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

Cool, I've folded the change in, thanks! Ethan On Fri, May 22, 2015 at 10:08 AM, Daniele Di Proietto wrote: > I was running an old version, getting it from the upstream repo helped, > thanks! > > I manage to compile everything without warnings if I apply the following > incremental > > ---8<---

Re: [ovs-dev] [PATCH v7] netdev-dpdk: add dpdk vhost-user ports

On Fri, May 22, 2015 at 04:40:00PM +0100, 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 > th

Re: [ovs-dev] [PATCH V2] ovn: Change type of 'chassis' column in 'Binding' table.

Thx, applied to master, On Fri, May 22, 2015 at 12:34 PM, Ben Pfaff wrote: > Fine with me. > > On Thu, May 21, 2015 at 10:02:12AM -0700, Alex Wang wrote: > > How about adding an inline function like this?~ > > > > On Thu, May 21, 2015 at 10:04 AM, Alex Wang wrote: > > > > > This commit changes

Re: [ovs-dev] [PATCH V2] ovn: Change type of 'chassis' column in 'Binding' table.

sorry, i mean applied to ovn branch, On Fri, May 22, 2015 at 1:38 PM, Alex Wang wrote: > Thx, applied to master, > > On Fri, May 22, 2015 at 12:34 PM, Ben Pfaff wrote: > >> Fine with me. >> >> On Thu, May 21, 2015 at 10:02:12AM -0700, Alex Wang wrote: >> > How about adding an inline function li

Re: [ovs-dev] [PATCH] ovn-northd: Add appctl support.

Thx, applied to ovn, On Fri, May 22, 2015 at 12:10 PM, Ben Pfaff wrote: > On Thu, May 21, 2015 at 03:20:48PM -0700, Alex Wang wrote: > > This commit adds appctl support to ovn-northd and the 'exit' command. > > The 'exit' command can be used in autotest for stopping the ovn-northd. > > > > Signe

Re: [ovs-dev] [PATCH] datapath-windows: Stateless TCP Tunnelling protocol - Initial implementation

hi Eitan, Thanks for this patch. Looks good overall. I had a few minor comments which I have inlined. Also, please update he instructions in INSTALL.Windows.md about the configuration steps for setting STT tunnels and testing them. thanks, -- Nithin > On May 7, 2015, at 6:08 PM, Eitan Eliah

Re: [ovs-dev] [PATCH v7] netdev-dpdk: add dpdk vhost-user ports

Pravin has been handling this on our end, so he'll do it. Btw how does this perform out of curiosity? Ethan On Fri, May 22, 2015 at 1:33 PM, Flavio Leitner wrote: > On Fri, May 22, 2015 at 04:40:00PM +0100, Ciara Loftus wrote: >> This patch adds support for a new port type to the userspace >> d

Re: [ovs-dev] [PATCH v7] netdev-dpdk: add dpdk vhost-user ports

On Fri, May 22, 2015 at 01:52:22PM -0700, Ethan Jackson wrote: > Pravin has been handling this on our end, so he'll do it. > > Btw how does this perform out of curiosity? These are devel numbers after doing some basic tuning: 64bytes, 13.3~13.7Mpps OVS+DPDK forwarding between 2 DPDK ports. (no v

[ovs-dev] dreaming of f#cking tonight?

Send me a F#ckFriends request so we can hook up My usename is VeronicaWet09 http://hop.kz/dobG";>my profile is here IM ONLINE ___ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev

[ovs-dev] [PATCH 2/2] odp-util: Format recirc action ID in hex.

The match is in hex, this makes it more consistent. Signed-off-by: Joe Stringer --- lib/odp-util.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/odp-util.c b/lib/odp-util.c index 962b84b..6e2a39d 100644 --- a/lib/odp-util.c +++ b/lib/odp-util.c @@ -495,7 +495,7 @@ forma

[ovs-dev] [PATCH 1/2] Revert "ovs-ofctl: Always prints recirc_id in decimal"

This reverts commit fa7de3220ce79ffe293d84cc9c98b1a88980273b. As there is the potential for this field to be maskable in future, and the dpctl "-m" output prints a mask for it, return it to hexadecimal. The next patch will make this consistent to the recirc action by making the action print the re

Re: [ovs-dev] [PATCH 2/2] odp-util: Format recirc action ID in hex.

On Fri, May 22, 2015 at 2:28 PM, Joe Stringer wrote: > The match is in hex, this makes it more consistent. > > Signed-off-by: Joe Stringer > --- > lib/odp-util.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/lib/odp-util.c b/lib/odp-util.c > index 962b84b..6e2a39d 1006

Re: [ovs-dev] [PATCH 2/2] odp-util: Format recirc action ID in hex.

On 22 May 2015 at 14:34, Andy Zhou wrote: > On Fri, May 22, 2015 at 2:28 PM, Joe Stringer wrote: >> The match is in hex, this makes it more consistent. >> >> Signed-off-by: Joe Stringer >> --- >> lib/odp-util.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/lib/odp-

Re: [ovs-dev] [PATCH v7] netdev-dpdk: add dpdk vhost-user ports

On Fri, May 22, 2015 at 1:33 PM, Flavio Leitner wrote: > On Fri, May 22, 2015 at 04:40:00PM +0100, 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 pro

Re: [ovs-dev] [PATCH 1/3] extract-ofp-fields: Fix most pep8 style issues.

On 20 May 2015 at 14:33, YAMAMOTO Takashi wrote: >> Signed-off-by: Joe Stringer > > Acked-by: YAMAMOTO Takashi Thanks, I applied the first two patches of this series to master. The third patch has a v2 here: http://openvswitch.org/pipermail/dev/2015-May/055439.html

Re: [ovs-dev] [PATCH] dpctl: Don't print UFID if not present.

On 22 May 2015 at 11:41, Ben Pfaff wrote: > On Fri, May 22, 2015 at 10:32:53AM -0700, Joe Stringer wrote: >> With verbose dpctl, if userspace runs against an older kernel, every >> entry will have "ufid:" at the beginning. This is unnecessary and >> introduces an additional format for scripts to p

Re: [ovs-dev] [PATCH 2/2] odp-util: Format recirc action ID in hex.

On Fri, May 22, 2015 at 2:38 PM, Joe Stringer wrote: > On 22 May 2015 at 14:34, Andy Zhou wrote: >> On Fri, May 22, 2015 at 2:28 PM, Joe Stringer wrote: >>> The match is in hex, this makes it more consistent. >>> >>> Signed-off-by: Joe Stringer >>> --- >>> lib/odp-util.c | 2 +- >>> 1 file cha

Re: [ovs-dev] FW: performance issue with ovs + dpdk2.0 with vhost

On Fri, May 15, 2015 at 02:07:07PM +, Rao, Ravi wrote: > > Hi All, >I am trying to get a Guest VM connected to the dpdkvhost >interface on a host which has the ovs running from the latest ovs >git and dpdk2.0. Looks like I am missing something as 95% of >traffic is not getting

[ovs-dev] vhost user cannot recv any packet from VM

Hi, I am using the latest ovs code base to try vhost user based on Ciara Loftus's patch ("[PATCH v4] netdev-dpdk: add dpdk vhost-user ports" which is not in ovs upstream yet). things go smoothly during VM bootstrap, as I can see following logs: VHOST_CONFIG: new virtio connection is 47 VHOST_C

Re: [ovs-dev] [PATCH v7] netdev-dpdk: add dpdk vhost-user ports

On Fri, May 22, 2015 at 8:40 AM, 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-