Hi All
As mentioned before, I am using a repo for DPDK patch merging.
The repo is here:
https://github.com/darball/ovs/
There are still some outstanding patches from Bhanu that have not completed
review yet:
util: Add PADDED_MEMBERS_CACHELINE_MARKER macro to mark cachelines.- Bhanu
packets: Reo
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Timothy M. Redaelli
> Sent: Tuesday, August 1, 2017 11:37 AM
> To: Weglicki, MichalX ; d...@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v4] Update relevant artifacts to
Hi Antonio
I noticed a problem in final testing and I made a small incremental change.
Pls let me know if this is ok ?
Thanks Darrell
diff --git a/lib/dpctl.c b/lib/dpctl.c
index 8eb0671..59f09b7 100644
--- a/lib/dpctl.c
+++ b/lib/dpctl.c
@@ -1518,8 +1518,7 @@ dpctl_ct_bkts(int argc, const char
> On Aug 1, 2017, at 12:09 PM, Ben Pfaff wrote:
>
> Signed-off-by: Ben Pfaff
I usually hold off on putting the changes in "debian/changelog", since NEWS
often continues to grow post-branch, and it's easy to miss updates. However, I
don't know how often people look at that file, so it may no
Hi, Ben
Ericsson, Redhat and Intel has thought OVS 2.8 would include NSH support, there
are some projects depending on this, for example, Opendaylight (Openflowplugin,
SFC and NetVirt), OPNFV and Openstack (networking-sfc).
This email is long and easy to make people misunderstand :-) my readout
On Wed, Aug 2, 2017 at 1:18 AM, Russell Bryant wrote:
> On Tue, Aug 1, 2017 at 3:26 PM, Han Zhou wrote:
> >
> >
> > On Tue, Aug 1, 2017 at 9:19 AM, Russell Bryant wrote:
> >>
> >> Add native support for active-standby HA in ovn-northd by having each
> >> instance attempt to acquire an OVSDB loc
Maybe
encap(nsh(md_type=2,tlv(0x1000,10,0x12345678),tlv(0x2000,20,0xfedcba9876543210))
is better, the below are possible formats, I think they are all ok, what do
you think about them?
encap(nsh)
encap(nsh())
encap(nsh(md_type=1))
encap(nsh(md_type=2,tlv(0x1000,10,0x12345678),tlv(0x2000,20,0xf
On Tue, Aug 01, 2017 at 07:53:18PM -0400, Russell Bryant wrote:
> On Tue, Aug 1, 2017 at 6:47 PM, Ben Pfaff wrote:
> > Hello everyone. Last week I spent an afternoon with the OVS-DPDK folks
> > at Intel in Shannon, Ireland. One of the ideas that came out of it was
> > a proposal for the contribu
On Tue, Aug 1, 2017 at 6:47 PM, Ben Pfaff wrote:
> Hello everyone. Last week I spent an afternoon with the OVS-DPDK folks
> at Intel in Shannon, Ireland. One of the ideas that came out of it was
> a proposal for the contributors at each company to publish, at the
> beginning of each release cycl
On 1 August 2017 at 15:49, Eric Garver wrote:
> On Tue, Aug 01, 2017 at 01:56:44PM -0700, Joe Stringer wrote:
>> On 1 August 2017 at 13:56, Joe Stringer wrote:
>> > On 29 July 2017 at 21:58, Roi Dayan wrote:
>> >> Since it's an error but also will always occur in older kernels
>> >> log the mess
On 1 August 2017 at 13:57, Joe Stringer wrote:
> On 28 July 2017 at 05:43, Simon Horman wrote:
>> On Thu, Jul 27, 2017 at 02:40:02PM +0300, Roi Dayan wrote:
>>> When we skip adding a port using rtnetlink and not because of an error we
>>> need to return EOPNOTSUPP to avoid logging an error messag
On 1 August 2017 at 15:47, Eric Garver wrote:
> When VXLAN-GPE was introduced we added IFLA_VXLAN_GPE to the policy
> parsing, but did not mark it as optional. The kernel only returns this
> netlink attribute if it's actually configured.
>
> This also adds a missing entry for IFLA_VXLAN_GBP. Appar
Bonjour cherJ'ai une proposition d'affaires qui sera avantageuse pour nous
deux, et je vous indemniserai avec le pourcentage à la conclusion finale. Si
vous êtes intéressé s'il vous plaît me répondre en arrière afin de vous fournir
plus de détails sur la façon dont nous pouvons procéder plus loi
encap(hdr=nsh,nsh(md_type=2,tlv(0x1000,10,0x12345678),tlv(0x2000,20,0xfedcba9876543210))
On Tue, Aug 01, 2017 at 10:53:27PM +, Yang, Yi Y wrote:
> Ben, because we're considering to cover NSH md type 2 case, for NSH TLV, now
> we provide it by the below way.
>
> encap(hdr=nsh,prop(class=nsh,t
An action has an embedded length. For nx_action_encap, if the embedded
length ->len is longer than sizeof(struct nx_action_encap), then
properties follow struct nx_action_encap until the length has been
exhausted.
On Tue, Aug 01, 2017 at 11:06:21PM +, Yang, Yi Y wrote:
> About why we need n_p
Le recordamos que nuestras Pólizas de Capacitación son una excelente opción
para capacitarse desde su oficina, sin gastos de traslado o viáticos y con
expertos de primer nivel que le darán a sus colaboradores la actualización que
necesitan para ejercer sus puestos de la manera más competitiva.
About why we need n_props in nx_action_encap, I added this for Opendaylight to
deserialize the wire format from OVS, n_props can clearly tell Opendaylight if
there is any property. Otherwise how do we check if there is a property
following?
-Original Message-
From: Ben Pfaff [mailto:b..
Got it, thanks a lot.
-Original Message-
From: Ben Pfaff [mailto:b...@ovn.org]
Sent: Wednesday, August 2, 2017 6:51 AM
To: Yang, Yi Y
Cc: Zoltán Balogh ; d...@openvswitch.org
Subject: Re: [ovs-dev] [PATCH v3 1/2] OF support and translation of generic
encap and decap
I don't understand
Ben, because we're considering to cover NSH md type 2 case, for NSH TLV, now we
provide it by the below way.
encap(hdr=nsh,prop(class=nsh,type=md_type,val=2),prop(class=nsh,type=tlv,val(0x1000,10,0x12345678)),prop(class=nsh,type=tlv,val(0x2000,20,0xfedcba9876543210)))
Can you help provide a form
I don't understand your explanation. decode_NXAST_RAW_DECAP() converts
an OpenFlow action in wire format into OVS internal format. The
OpenFlow action that it contains might have been generated by anything
(for example, a controller). parse_ENCAP() doesn't have anything to do
with it as far as I
On Tue, Aug 01, 2017 at 01:56:44PM -0700, Joe Stringer wrote:
> On 1 August 2017 at 13:56, Joe Stringer wrote:
> > On 29 July 2017 at 21:58, Roi Dayan wrote:
> >> Since it's an error but also will always occur in older kernels
> >> log the message with level warning instead of info.
> >>
> >> Sig
Hello everyone. Last week I spent an afternoon with the OVS-DPDK folks
at Intel in Shannon, Ireland. One of the ideas that came out of it was
a proposal for the contributors at each company to publish, at the
beginning of each release cycle, a list of what features and other
contributions they we
When VXLAN-GPE was introduced we added IFLA_VXLAN_GPE to the policy
parsing, but did not mark it as optional. The kernel only returns this
netlink attribute if it's actually configured.
This also adds a missing entry for IFLA_VXLAN_GBP. Apparently we have no
system-traffic test coverage there.
Fi
Ben, thank you for your patch, but from my understanding, parse_ENCAP has
ensured it is impossible to have any property for decap, so I'm not sure in
what case this will happened.
-Original Message-
From: Ben Pfaff [mailto:b...@ovn.org]
Sent: Tuesday, August 1, 2017 11:26 PM
To: Yang, Y
Guru, could you please apply this patch.
Thanks,
Shashank
From: Anand Kumar
Sent: Thursday, July 27, 2017 4:48:02 PM
To: Shashank Ram; d...@openvswitch.org
Subject: Re: [ovs-dev] [PATCH v2] datapath-windows: Refactor
OvsCreateNewNBLsFromMultipleNBs
Thank
From: ovs-dev-boun...@openvswitch.org on
behalf of Anand Kumar
Sent: Tuesday, August 1, 2017 3:01 PM
To: d...@openvswitch.org
Subject: [ovs-dev] [PATCH] datapath-windows: Fix conntrack lookups for
reversed keys
From: Sairam Venugopal
The conntrack
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 the
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 pre-pr
This commit changes the service file from static configuration to an
autogenerated file, produced during the build. This will be relevant in a
future commit.
Signed-off-by: Aaron Conole
---
rhel/.gitignore | 1 +
rhel/automake.mk
When intermediary files are generated, the destination directory is assumed
to exist. This has worked so far because most files are built prior to
the dist-packaging step. However, any files which require rebuild after
the packaging step may end up in failure if the output directory is not
availa
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.
To do this in a way that doesn't impact e
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
root:r
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 information.
From: Sairam Venugopal
The conntrack table needs to be queried for entries in either directions
to determine if the packet is in forward direction or reply direction.
The current behavior ends up reversing the incoming packet's 5-Tuple for
every entry in the loop instead of doing it only once.
On 28 July 2017 at 05:43, Simon Horman wrote:
> On Thu, Jul 27, 2017 at 02:40:02PM +0300, Roi Dayan wrote:
>> When we skip adding a port using rtnetlink and not because of an error we
>> need to return EOPNOTSUPP to avoid logging an error message.
>>
>> Fixes: 2fd3d5eda508 ("dpif-netlink-rtnl: Sup
On 29 July 2017 at 22:01, Roi Dayan wrote:
> From: Paul Blakey
>
> Use sysconf(_SC_CLK_TCK) to read run time "number of clock ticks per
> second" and use that to convert ticks to msecs.
> This is how iproute does the conversion when parsing tc filters.
> The system call is done only once.
>
> Sig
On 1 August 2017 at 13:56, Joe Stringer wrote:
> On 29 July 2017 at 21:58, Roi Dayan wrote:
>> Since it's an error but also will always occur in older kernels
>> log the message with level warning instead of info.
>>
>> Signed-off-by: Roi Dayan
>> ---
>
> When I run this patch on a fedora system
On 29 July 2017 at 21:58, Roi Dayan wrote:
> Since it's an error but also will always occur in older kernels
> log the message with level warning instead of info.
>
> Signed-off-by: Roi Dayan
> ---
When I run this patch on a fedora system with kernel 4.8 and
out-of-tree modules, when running the
On Tue, Aug 01, 2017 at 06:49:53PM +, Darrell Ball wrote:
>
> -Original Message-
> From: on behalf of Ben Pfaff
> Date: Tuesday, August 1, 2017 at 10:23 AM
> To: "d...@openvswitch.org"
> Subject: [ovs-dev] RFC: Let's branch for 2.8 today
>
> I'd like to create branch-2.8 today
On Tue, Aug 1, 2017 at 1:23 PM, Ben Pfaff wrote:
> I'd like to create branch-2.8 today in preparation for releasing OVS 2.8
> later in August. I sent the patch that would lead off the branch a few
> minutes ago:
> https://mail.openvswitch.org/pipermail/ovs-dev/2017-August/336544.html
>
>
On Tue, Aug 1, 2017 at 3:26 PM, Han Zhou wrote:
>
>
> On Tue, Aug 1, 2017 at 9:19 AM, Russell Bryant wrote:
>>
>> Add native support for active-standby HA in ovn-northd by having each
>> instance attempt to acquire an OVSDB lock. Only the instance of
>> ovn-northd that currently holds the lock w
On Tue, Aug 1, 2017 at 9:19 AM, Russell Bryant wrote:
>
> Add native support for active-standby HA in ovn-northd by having each
> instance attempt to acquire an OVSDB lock. Only the instance of
> ovn-northd that currently holds the lock will make active changes to
> the OVN databases.
>
> Signed-
-Original Message-
From: on behalf of Ben Pfaff
Date: Tuesday, August 1, 2017 at 10:23 AM
To: "d...@openvswitch.org"
Subject: [ovs-dev] RFC: Let's branch for 2.8 today
I'd like to create branch-2.8 today in preparation for releasing OVS 2.8
later in August. I sent the patch th
On Tue, Aug 01, 2017 at 10:24:47AM -0700, Joe Stringer wrote:
> On 1 August 2017 at 07:55, Ben Pfaff wrote:
> > On Mon, Jul 31, 2017 at 05:16:11PM -0700, Joe Stringer wrote:
> >> Clang complains:
> >>
> >> In file included from include/openvswitch/cxxtest.cc:11:0:
> >> ../include/openvswitch/ofpbu
On 1 August 2017 at 07:55, Ben Pfaff wrote:
> On Mon, Jul 31, 2017 at 05:16:11PM -0700, Joe Stringer wrote:
>> Clang complains:
>>
>> In file included from include/openvswitch/cxxtest.cc:11:0:
>> ../include/openvswitch/ofpbuf.h: In function ‘ofpbuf
>> ofpbuf_const_initializer(const void*, size_t)
I'd like to create branch-2.8 today in preparation for releasing OVS 2.8
later in August. I sent the patch that would lead off the branch a few
minutes ago:
https://mail.openvswitch.org/pipermail/ovs-dev/2017-August/336544.html
According to our published schedule, we should have create br
Signed-off-by: Ben Pfaff
---
This is intended to be the first commit on a newly created branch-2.8.
NEWS | 2 +-
configure.ac | 2 +-
debian/changelog | 88 ++--
3 files changed, 81 insertions(+), 11 deletions(-)
diff --git a
On 07/31/2017 04:56 PM, Darrell Ball wrote:
Hi Greg
Since you would be looking at the DPDK documentation these days; could you help
review this patch ?
Thanks Darrell
The patch looks good to me Darrell. Replying to your forward since my email
client seems to have dumped the original patch
On Fri, Jul 28, 2017 at 3:02 PM, Timothy Redaelli wrote:
> Replace systemctk with systemctl
>
> Signed-off-by: Timothy Redaelli
> ---
> rhel/README.RHEL.rst | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
Thanks, applied to master.
--
Russell Bryant
___
Add native support for active-standby HA in ovn-northd by having each
instance attempt to acquire an OVSDB lock. Only the instance of
ovn-northd that currently holds the lock will make active changes to
the OVN databases.
Signed-off-by: Russell Bryant
---
NEWS| 1 +
ovn
> -Original Message-
> From: O Mahony, Billy
> Sent: Monday, July 31, 2017 5:22 PM
> To: Fischetti, Antonio ; d...@openvswitch.org
> Subject: RE: [ovs-dev] [PATCH v2 2/5] dpif-netdev: Avoid reading RSS hash when
> EMC is disabled.
>
> Hi Antonio,
>
> This is patch is definitely simpler
Rxqs consumed processing cycles are used to improve the balance
of how rxqs are assigned to pmds. Currently some reconfiguration
is needed to perform a reassignment.
Add an ovs-appctl command to perform a new assignment in order
to balance based on the latest rxq processing cycle information.
Not
Up to his point rxqs are sorted by processing cycles they
consumed and assigned to pmds in a round robin manner.
Ian pointed out that on wrap around the most loaded pmd will be
the next one to be assigned an additional rxq and that it would be
better to reverse the pmd order when wraparound occurs
Previously rxqs were assigned to pmds by round robin in
port/queue order.
Now that we have the processing cycles used for existing rxqs,
use that information to try and produced a better balanced
distribution of rxqs across pmds. i.e. given multiple pmds, the
rxqs which have consumed the largest a
Count the cycles used for processing an rxq during the
pmd rxq interval. As this is an in flight counter and
pmds run independently, also store the total cycles used
during the last full interval.
Signed-off-by: Kevin Traynor
---
lib/dpif-netdev.c | 65 +++
Add two counters to dp_netdev_rxq which will be used for storing the
processing cycles of an rxq. Processing cycles will be stored in reference
to a defined interval. One counter is used for storing cycles during the
current in progress interval, while the other is used to store the cycles
of the l
Soon we will want to store processing cycle counts in the dp_netdev_rxq,
so use that as a basis for the polled_queue that pmd_thread_main uses.
Signed-off-by: Kevin Traynor
---
lib/dpif-netdev.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/lib/dpif-netdev.c b/li
For the DPDK datapath, by default rxqs are assigned to available pmds
in round robin order with no weight or priority.
It can happen that some very busy queues are handled by one pmd which
does not have enough cycles to prevent packets being dropped on them.
While at the same time another pmd whic
On Tue, Aug 01, 2017 at 08:06:59PM +0800, Yi Yang wrote:
> From: Jan Scheurich
>
> This commit adds support for the OpenFlow actions generic encap
> and decap (as specified in ONF EXT-382) to the OVS control plane.
>
> CLI syntax for encap action with properties:
> encap(hdr=)
> encap(hdr=,
On 07/22/2017 03:52 PM, Stokes, Ian wrote:
>> Up to his point rxqs are sorted by processing cycles they consumed and
>> assigned to pmds in a round robin manner.
>>
>> Ian pointed out that on wrap around the most loaded pmd will be the next
>> one to be assigned an additional rxq and that it would
On 07/22/2017 03:52 PM, Stokes, Ian wrote:
>> Previously rxqs were assigned to pmds by round robin in port/queue order.
>>
>> Now that we have the processing cycles used for existing rxqs, use that
>> information to try and produced a better balanced distribution of rxqs
>> across pmds. i.e. given
On 07/22/2017 03:51 PM, Stokes, Ian wrote:
>> Add two counters to dp_netdev_rxq which will be used for storing the
>> processing cycles of an rxq. Processing cycles will be stored in reference
>> to a defined interval. One counter is used for storing cycles during the
>> current in progress interva
On 07/22/2017 03:50 PM, Stokes, Ian wrote:
>> So far the interval was only used for dpcls optimization.
>> Soon, we will use it for storing rxq cycles so make the names more
>> generic. Also, set the interval regardless of whether dpcls optimization
>> has occurred, as the optimization interval wil
>On 28.07.2017 10:20, Darrell Ball wrote:
>> I have not tested yet
>>
>> However, I would have expected something max latency config. to be
>specific to netdev-dpdk port types
>
>IMHO, if we can make it generic, we must make it generic.
>
>[Darrell]
>The first question I ask mys
On Tue, Aug 01, 2017 at 08:26:07AM -0700, Ben Pfaff wrote:
> On Tue, Aug 01, 2017 at 12:32:20PM +, Yang, Yi Y wrote:
> > #2.
> > [Ben] I suspect that decode_NXAST_RAW_DECAP() should report an error if
> > properties are present, since it doesn't support properties.
> >
> > [Yi] It is impossib
>This patch-set inspired by [1] from Bhanuprakash Bodireddy.
>Implementation of [1] looks very complex and introduces many pitfalls for
>later code modifications like possible packet stucks.
>
>This version targeted to make simple and flexible output packet batching on
>higher level without introdu
On Tue, Aug 01, 2017 at 12:32:20PM +, Yang, Yi Y wrote:
> #2.
> [Ben] I suspect that decode_NXAST_RAW_DECAP() should report an error if
> properties are present, since it doesn't support properties.
>
> [Yi] It is impossible.
What is impossible? It is easy to detect that properties are pre
On Tue, Aug 01, 2017 at 12:04:55AM +, Yang, Yi Y wrote:
> Hi, Ben
>
> Thank you so much for your great review and great comments, I'll do changes
> per your comments and post next version because Zoltan and Jan are taking
> vacation. I explained your concerns
>
> [Ben] The string format o
Dear Ben,
Thanks. This make me clear.
Ted
Sent from Mail for Windows 10
From: Ben Pfaff
Sent: Tuesday, August 1, 2017 10:53 PM
To: ted.y.li...@gmail.com
Cc: ovs-dev@openvswitch.org
Subject: Re: [ovs-dev] How to tell which code is for userspace
On Tue, Aug 01, 2017 at 10:01:31PM +0800, ted.y.li
On Mon, Jul 31, 2017 at 05:16:11PM -0700, Joe Stringer wrote:
> Clang complains:
>
> In file included from include/openvswitch/cxxtest.cc:11:0:
> ../include/openvswitch/ofpbuf.h: In function ‘ofpbuf
> ofpbuf_const_initializer(const void*, size_t)’:
> ../include/openvswitch/ofpbuf.h:107:5: warning
On Tue, Aug 01, 2017 at 10:01:31PM +0800, ted.y.li...@gmail.com wrote:
> I am reading ovs 2.7.90 code, do you have a quick way to tell which code is
> for userspace and which for kernel?
Linux kernel code is in the datapath/ directory.
Windows kernel code is in the datapath-windows/ directory.
Th
Thanks.
But there is a function in datapath/vport.c, as below:
struct vport *ovs_vport_alloc(int priv_size, const struct vport_ops *ops,
const struct vport_parms *parms)
{
struct vport *vport;
size_t alloc_size;
alloc_size = sizeof(struct vport);
On 01.08.2017 17:19, Ilya Maximets wrote:
> On 01.08.2017 15:46, Keshav Gupta wrote:
>> Hi Ben/Ilya
>> Mainly patch do the following
>> 1) Call the dpdk rx api(rte_eth_rx_burst) only when ovs netdev state is UP
>> 2) Now the rte_eth_dev_stop/ rte_eth_dev_start calls are removed in
>> upda
Ben Pfaff writes:
> On Mon, Jul 31, 2017 at 04:44:10PM -0400, Russell Bryant wrote:
>> On Mon, Jul 31, 2017 at 4:33 PM, Ben Pfaff wrote:
>> > On Mon, Jul 31, 2017 at 01:05:32PM -0700, Ben Pfaff wrote:
>> >> On Mon, Jul 31, 2017 at 03:50:59PM -0400, Russell Bryant wrote:
>> >> > On Sun, Jul 30, 2
Currently, the dpdk-vhost side in ovs doesn't support tcp/udp tx cksum.
So L4 packets's cksum were calculated in VM side but performance is not
good.
Implementing tcp/udp tx cksum in ovs-dpdk side improves throughput and
makes virtio-net frontend-driver support NETIF_F_SG as well
Signed-off-by: Zh
On 01.08.2017 15:46, Keshav Gupta wrote:
> Hi Ben/Ilya
> Mainly patch do the following
> 1) Call the dpdk rx api(rte_eth_rx_burst) only when ovs netdev state is UP
> 2) Now the rte_eth_dev_stop/ rte_eth_dev_start calls are removed in updating
> the flags
>
>
> This fixes the core du
Hi Antonio,
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of antonio.fische...@intel.com
> Sent: Wednesday, July 19, 2017 5:05 PM
> To: d...@openvswitch.org
> Subject: [ovs-dev] [PATCH v2 5/5] dp-packet: Use memcpy on dp_
On Tue, Aug 1, 2017 at 10:01 AM, wrote:
> Dear all,
>
> I am reading ovs 2.7.90 code, do you have a quick way to tell which code is
> for userspace and which for kernel?
The ovs/datapath/linux/ directory is the Linux kernel datapath code.
The rest is userspace (with the exception of the Windows
Dear all,
I am reading ovs 2.7.90 code, do you have a quick way to tell which code is for
userspace and which for kernel?
Br,
Ted
Sent from Mail for Windows 10
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/o
On Tue, Aug 1, 2017 at 8:06 AM, Numan Siddique wrote:
>
>
> On Tue, Aug 1, 2017 at 4:25 PM, Numan Siddique wrote:
>>
>>
>>
>> On Tue, Aug 1, 2017 at 5:00 AM, Han Zhou wrote:
>>>
>>> On Mon, Jul 31, 2017 at 1:53 PM, Russell Bryant wrote:
>>> >
>>> > I wanted to share the idea before I code it to
Hi All,
This proposal is an attempt to make a more general solution to the same issue
of EMC thrashing addressed by
https://mail.openvswitch.org/pipermail/ovs-dev/2017-July/335940.html. That patch
proposes that when the EMC is overloaded that recirculated packets are neither
inserted into the EMC
Hi Ben/Ilya
Mainly patch do the following
1) Call the dpdk rx api(rte_eth_rx_burst) only when ovs netdev state is UP
2) Now the rte_eth_dev_stop/ rte_eth_dev_start calls are removed in updating
the flags
This fixes the core dump issue but I see below side effect with this patch:
Wit
Hi, Ben
I have posted v4, they are:
https://mail.openvswitch.org/pipermail/ovs-dev/2017-August/336504.html
https://mail.openvswitch.org/pipermail/ovs-dev/2017-August/336505.html
https://mail.openvswitch.org/pipermail/ovs-dev/2017-August/336506.html
I have fixed most of issues you commented, I ne
From: Zoltan Balogh
- Checking decap() prerequisits.
- Encap/decap VLAN tagged Ethernet frames.
- Send L3 packet over patch port.
- Output L2/L3 packet to ports with different packet_type properties.
Signed-off-by: Zoltan Balogh
Suggested-by: Jan Scheurich
---
tests/packet-type-aware.
From: Jan Scheurich
This commit adds support for the OpenFlow actions generic encap
and decap (as specified in ONF EXT-382) to the OVS control plane.
CLI syntax for encap action with properties:
encap(hdr=)
encap(hdr=,
prop(class=,type=,val=),
prop(class=,type=,val()))
CLI s
This series is a continuation of other patch series initiated by Jan Scheurich
before.
The main purpose of this series is to add support for the OpenFlow actions
generic encap and decap (ONF EXT-382) to the OVS control plane. It implements
a skeleton for translation of generic encap and decap ac
On Tue, Aug 1, 2017 at 4:25 PM, Numan Siddique wrote:
>
>
> On Tue, Aug 1, 2017 at 5:00 AM, Han Zhou wrote:
>
>> On Mon, Jul 31, 2017 at 1:53 PM, Russell Bryant wrote:
>> >
>> > I wanted to share the idea before I code it to see if it makes sense.
>> > I imagine the patch would be small, though
> -Original Message-
> From: O Mahony, Billy
> Sent: Monday, July 31, 2017 4:33 PM
> To: Fischetti, Antonio ; d...@openvswitch.org
> Subject: RE: [ovs-dev] [PATCH v2 1/5] dpif-netdev: move pkt metadata init out
> of emc_processing.
>
> There is also a reference to md_is_valid is the comm
On Tue, Aug 1, 2017 at 5:00 AM, Han Zhou wrote:
> On Mon, Jul 31, 2017 at 1:53 PM, Russell Bryant wrote:
> >
> > I wanted to share the idea before I code it to see if it makes sense.
> > I imagine the patch would be small, though.
> >
> > We currently provide HA for ovn-northd by using Pacemaker
Hi Antonio,
Unfortunately I think the performance deltas of this here probably need to be
re-worked given the bug discovered & fixed in EMC Insertion algorithm here
which according to the patch notes will significantly reduce EMC contention for
a given number of flows.
https://mail.openvswitch
On 07/25/2017 02:34 PM, Michal Weglicki wrote:
Upgrading to DPDK 17.05.1 stable release adds new
significant features relevant to OVS, including,
but not limited to:
- tun/tap PMD,
- VFIO hotplug support,
- Generic flow API.
Following changes are applied:
- netdev-dpdk: Changes required by DPDK
91 matches
Mail list logo