___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Hi Jiri,
On Tue, May 10, 2016 at 02:06:18PM +0200, Jiri Benc wrote:
> On Mon, 9 May 2016 17:18:20 +0900, Simon Horman wrote:
> > On Fri, May 06, 2016 at 11:35:04AM +0200, Jiri Benc wrote:
> > > In addition, we should check whether mac_len > 0 and in such case,
> > > change skb->protocol to ETH_P_T
For traffic from physical space to virtual space we need DNAT.
The DNAT happens in the gateway router and reaches the logical
port. The return traffic should be SNATed.
Traffic originating in virtual space heading to physical space
should be SNATed.
East-west traffic with the public destination I
OVS NAT currently cannot do snat and dnat in the same zone.
So we need two zones per gateway router.
Signed-off-by: Gurucharan Shetty
---
No v1. Newly added as part of v2.
---
ovn/controller/ovn-controller.c | 50 +--
ovn/controller/ovn-controller.h |2 +
We currently allocate conntrack zones in binding.c. It fits
in nicely there because we currently only allocate conntrack
zones to logical ports and binding.c is where we figure out
the local ones.
An upcoming commit needs conntrack zone allocation for routers
in a gateway. For that reason, this co
Currently OVN has distributed switches and routers. When a packet
exits a container or a VM, the entire lifecycle of the packet
through multiple switches and routers are calculated in source
chassis itself. When the destination endpoint resides on a different
chassis, the packet is sent to the othe
Currently we can connect routers via "peer"ing. This limits
the number of routers that can be connected with each other
directly to 2.
One of the design goals for L3 Gateway is to be able to
have multiple gateways (each with their own router)
connected to a distributed router via a switch.
With t
Hi Jiri,
On Wed, May 11, 2016 at 10:50:09AM +0900, Simon Horman wrote:
[...]
> > > Its possible that I've overlooked something but as things stand I think
> > > things look like this:
> > >
> > > * ovs_flow_key_extract() keys off dev->type and skb->protocol.
> > > * ovs_flow_key_extract() calls
On Wed, May 11, 2016 at 4:31 AM, Flavio Leitner wrote:
> On Tue, May 10, 2016 at 10:31:19AM +0800, Xiao Liang wrote:
>> On Tue, May 10, 2016 at 4:28 AM, Flavio Leitner wrote:
>> > On Sat, Apr 23, 2016 at 01:26:17PM +0800, Xiao Liang wrote:
>> >> Add new column "ofname" in Interface table to confi
Hi Jiri,
On Tue, May 10, 2016 at 02:01:06PM +0200, Jiri Benc wrote:
> On Mon, 9 May 2016 17:04:22 +0900, Simon Horman wrote:
> > It seems to be caused by the following:
> >
> > 1. __ipgre_rcv() calls skb_pop_mac_header() which
> >sets skb->mac_header to the skb->network_header.
> >
> > 2. __
On Tue, May 10, 2016 at 1:21 AM, Markos Chandras wrote:
> Commit e2f3178f0582 ("datapath: Add support for kernel 3.14.") added
> support for 3.14 kernels and a new OVS_GREP_IFELSE check for the
> "skg_get_hash" function in the process. "skb_get_hash" was introduced
> in the Linux kernel commit 395
On Tue, May 10, 2016 at 3:26 AM, Chandran, Sugesh
wrote:
>> -Original Message-
>> From: Jesse Gross [mailto:je...@kernel.org]
>> Sent: Friday, May 6, 2016 5:00 PM
>> To: Chandran, Sugesh
>> Cc: pravin shelar ; ovs dev
>> Subject: Re: [ovs-dev] [PATCH v2] tunneling: Improving tunneling
>>
The original message was received at Wed, 11 May 2016 07:43:50 +0700 from
28.185.172.64
- The following addresses had permanent fatal errors -
dev@openvswitch.org
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinf
Dear user dev@openvswitch.org, Mail server administrator of openvswitch.org
would like to let you know that:
We have received reports that your account has been used to send a large amount
of spam during the last week.
Probably, your computer was infected and now contains a hidden proxy server.
> On May 10, 2016, at 7:34 AM, Numan Siddique wrote:
>
> Recirculations due to NXT_RESUME are failing if the packet metadata is not
> restored prior to the packet execution.
>
> Reported-at: http://openvswitch.org/pipermail/dev/2016-May/070723.html
> Signed-Off-by: Numan Siddique
> ---
> ofpro
On Tue, May 10, 2016 at 1:39 PM, Justin Pettit wrote:
>
> > On May 7, 2016, at 9:21 AM, Darrell Ball wrote:
> >
> > This patch updates the vtep schema, vtep-ctl commands and vtep simulator
> > to support source node replication in addition to service node
> replication per
> > logical switch. T
Acked-by: Jarno Rajahalme
It seems to me that the UFID is still passed down alongside the key, assuming
that is ok.
Jarno
> On May 10, 2016, at 3:42 PM, Joe Stringer wrote:
>
> Windows datapath folks have reported instances where OVS userspace will
> pass down a flow_get request to the dat
Clang complains:
lib/netdev-dpdk.c:1860:1: error: mutex 'dev->mutex' is not locked on every path
through here [-Werror,-Wthread-safety-analysis]
}
^
lib/netdev-dpdk.c:1815:5: note: mutex acquired here
ovs_mutex_lock(&dev->mutex);
^
./include/openvswitch/thread.h:60:9: note: expanded f
Windows datapath folks have reported instances where OVS userspace will
pass down a flow_get request to the datapath using a UFID even though the
datapath has no support for UFIDs. Since commit e672ff9b4d22
("ofproto-dpif: Restore metadata and registers on recirculation."), if a
flow dump provides
On 10 May 2016 at 10:37, pravin shelar wrote:
> On Fri, May 6, 2016 at 9:23 AM, Ben Pfaff wrote:
>> On Mon, Apr 25, 2016 at 03:58:33PM -0700, Pravin B Shelar wrote:
>>> The neighbor entry expiry is only checked in dpif-poll
>>> event handler, But in absence of any event we could keep
>>> using ar
> On May 10, 2016, at 1:44 PM, Joe Stringer wrote:
>
> On 3 May 2016 at 18:31, Justin Pettit wrote:
>> Commit f2d105b5 (ofproto-dpif-xlate: xlate ct_{mark, label} correctly.)
>> introduced the ovs_u128_and() function. It directly takes ovs_u128
>> values as arguments instead of pointers to the
On 3 May 2016 at 18:31, Justin Pettit wrote:
> Commit f2d105b5 (ofproto-dpif-xlate: xlate ct_{mark, label} correctly.)
> introduced the ovs_u128_and() function. It directly takes ovs_u128
> values as arguments instead of pointers to them. As this is a bit more
> direct way to deal with 128-bit v
> On May 7, 2016, at 9:21 AM, Darrell Ball wrote:
>
> This patch updates the vtep schema, vtep-ctl commands and vtep simulator
> to support source node replication in addition to service node replication per
> logical switch. The default replication mode is service node as that was the
> only m
On Tue, May 10, 2016 at 10:31:19AM +0800, Xiao Liang wrote:
> On Tue, May 10, 2016 at 4:28 AM, Flavio Leitner wrote:
> > On Sat, Apr 23, 2016 at 01:26:17PM +0800, Xiao Liang wrote:
> >> Add new column "ofname" in Interface table to configure port name reported
> >> to controllers with OpenFlow pro
There is no particularly good reason to use our own Python JSON
serialization implementation when serialization can be done faster
with Python's built-in JSON library.
A few tests were changed due to Python's default JSON library
returning slightly more precise floating point numbers and returning
The pure Python in-tree JSON parser is *much* slower than the
in-tree C JSON parser. A local test parsing a 100Mb JSON file
showed the Python version taking 270 seconds. With the C wrapper,
it took under 4 seconds.
The C extension will be used automatically if it can be built. If
the extension fai
Adds support for using the in-tree C JSON parser in Python. It also
uses the built-in Python JSON parser for serialization. In a test of
parsing a 100Mb JSON file, this patch improves performance by 70x. This
version corrects the code formatting issues with the previous version.
I'm having trouble
Thanks for this patch, Mauricio.
More inline.
On Tue, May 10, 2016 at 09:47:12PM +0200, Mauricio Vásquez wrote:
> Adding in CC all the people involved.
> (I don't know why git just takes two)
>
> On Tue, May 10, 2016 at 9:11 PM, Mauricio Vasquez B <
> mauricio.vasquezber...@studenti.polito.it> w
On 5 May 2016 at 12:54, Jarno Rajahalme wrote:
> Thanks for figuring this out, Joe!
>
> Acked-by: Jarno Rajahalme
No problem, thanks for looking the series over. I applied this to master.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/
On 5 May 2016 at 13:05, Jarno Rajahalme wrote:
> Acked-by: Jarno Rajahalme
Thanks, applied.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On 5 May 2016 at 13:04, Jarno Rajahalme wrote:
> Acked-by: Jarno Rajahalme
Thanks, applied.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On Tue, May 10, 2016 at 11:02 AM, Guru Shetty wrote:
>
>
> On 9 May 2016 at 23:57, Mickey Spiegel wrote:
>
>> The big issue for me is your last comment:
>>
>> If you add bob2 on hv1, would bob2 be able to reach alice1?
>>
>> No; take foo1 which is already on hv1 for this example
Adding in CC all the people involved.
(I don't know why git just takes two)
On Tue, May 10, 2016 at 9:11 PM, Mauricio Vasquez B <
mauricio.vasquezber...@studenti.polito.it> wrote:
> In order to use dpdk ports in ovs they have to be bound to a DPDK
> compatible driver before ovs is started.
>
> Th
Hi Markos,
I sent a new version:
http://openvswitch.org/pipermail/dev/2016-May/070795.html
Thank you for your feedback.
On Tue, May 3, 2016 at 5:27 PM, Markos Chandras wrote:
> Hello,
>
> (a couple of minor things I noticed whilst looking at your patch)
> On 04/13/2016 03:08 PM, Mauricio Vasqu
In order to use dpdk ports in ovs they have to be bound to a DPDK
compatible driver before ovs is started.
This patch adds the possibility to hotplug (or hot-unplug) a device
after ovs has been started. The implementation adds an appctl command:
netdev-dpdk/port-clt
After the user attaches a new
hi dev,
As promised, the document you requested is attached
Regards,
Kara Barry
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
hi dev,
As promised, the document you requested is attached
Regards,
Vincenzo Lee
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
hi dev,
As promised, the document you requested is attached
Regards,
Liz Mcdaniel
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On 9 May 2016 at 23:57, Mickey Spiegel wrote:
> The big issue for me is your last comment:
>
> If you add bob2 on hv1, would bob2 be able to reach alice1?
>
> No; take foo1 which is already on hv1 for this example;
> if I try to send b/w foo1 and alice1, I have “no route to alice
Currently OVN has distributed switches and routers. When a packet
exits a container or a VM, the entire lifecycle of the packet
through multiple switches and routers are calculated in source
chassis itself. When the destination endpoint resides on a different
chassis, the packet is sent to the othe
On Fri, May 6, 2016 at 9:23 AM, Ben Pfaff wrote:
> On Mon, Apr 25, 2016 at 03:58:33PM -0700, Pravin B Shelar wrote:
>> The neighbor entry expiry is only checked in dpif-poll
>> event handler, But in absence of any event we could keep
>> using arp entry forever. This patch changes it to check
>> ex
This patch used userpsace tunneling mechanism for implementing
STT tunneling protocol.
For details about STT you can refer the draft:
https://tools.ietf.org/html/draft-davie-stt-07
Signed-off-by: Pravin B Shelar
---
lib/automake.mk | 2 +
lib/netdev-native-stt.c | 708 +
This patch adds support to segment large UDP based tunnel
packet. With this patch large packet generated by STT can
be forwarded over to GENEVE/VXLAN port.
Signed-off-by: Pravin B Shelar
---
lib/dp-packet-lso.c | 32 +++-
lib/dp-packet-lso.h | 1 +
lib/netdev
This patch adds support to segment large GRE packet. This does
means that there are two sets of headers for given packet for
segmentation. To get offset of inner packet outer_hlen member
is added to dp-packet.
Signed-off-by: Pravin B Shelar
---
lib/dp-packet-lso.c | 74 ++
This function can be used to build varius tunnel headers.
Signed-off-by: Pravin B Shelar
---
lib/netdev-native-tnl.c | 128 ++--
lib/netdev-native-tnl.h | 28 +--
lib/netdev-provider.h | 9 ++--
lib/netdev.c| 11 +++--
lib/net
STT uses TCP port so we need to filter traffic on basis of TCP
port numbers.
Signed-off-by: Pravin B Shelar
Acked-by: Jesse Gross
---
lib/tnl-ports.c | 82 +---
lib/tnl-ports.h | 4 +--
ofproto/tunnel.c | 8 --
3 files changed, 62 inse
Set and get functions for IP explicit congestion notification flag.
These function would be used by STT reassembly code.
Signed-off-by: Pravin B Shelar
Acked-by: Jesse Gross
---
lib/packets.c| 21 +
lib/packets.h| 7 +++
ofproto/tunnel.c | 6 +++---
3 files cha
STT can generate large packets. Following patch adds support to handle
packet which are larger than device MTU size.
To support such packets following members are added dp-packet structure.
type: type of packet, e.g tcp, tcpv6, geneve, etc.
mss: maximum segment size.
netdev can set the flags in net
Large segment support need to use this refactored function to
send individual segments.
Signed-off-by: Pravin B Shelar
Acked-by: Jesse Gross
---
lib/dpif-netdev.c | 41 ++---
1 file changed, 26 insertions(+), 15 deletions(-)
diff --git a/lib/dpif-netdev.c b/
Once datapath support large packets, we need to segment packet before
sending it to upcall. Refactoring this code make it bit cleaner.
Signed-off-by: Pravin B Shelar
Acked-by: Jesse Gross
---
lib/dpif-netdev.c | 128 +-
1 file changed, 70 inse
The tunnel header pop action can leak batch of packet
in case of error. Following patch fixex the error code path.
Signed-off-by: Pravin B Shelar
---
lib/dpif-netdev.c | 21 -
lib/netdev.c | 7 ---
lib/netdev.h | 2 +-
3 files changed, 13 insertions(+), 17 del
in case of error from netdev_push_header() batch of packets was not
freed. Following patch fixes this issue.
Signed-off-by: Pravin B Shelar
---
lib/dpif-netdev.c | 16 ++--
1 file changed, 10 insertions(+), 6 deletions(-)
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index f0d5
DPDK datapath operate on batch of packets. To pass the batch of
packets around we use packets array and count. Next patch needs
to associate meta-data with each batch of packets. So Introducing
a batch structure to make handling the metadata easier.
Signed-off-by: Pravin B Shelar
Acked-by: Jesse
Signed-off-by: Pravin B Shelar
Acked-by: Jesse Gross
---
lib/dp-packet.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/dp-packet.h b/lib/dp-packet.h
index 201fd14..4a8b5ab 100644
--- a/lib/dp-packet.h
+++ b/lib/dp-packet.h
@@ -516,7 +516,7 @@ dp_packet_reset_packet(stru
This scratchpad can be used by any layer to keep private data.
STT will use it for TCP reassembly state.
Signed-off-by: Pravin B Shelar
Acked-by: Jesse Gross
---
lib/dp-packet.h | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/lib/dp-packet.h b/lib/dp-packet.h
index 000
Next patch introduces new structure named packet_batch. So
I am renaming it to packet_batch_per_flow.
This does not change any functionality.
Signed-off-by: Pravin B Shelar
Acked-by: Jesse Gross
---
lib/dpif-netdev.c | 36 +++-
1 file changed, 19 insertions(+), 1
Following patch series adds support for STT. STT can generate large
packet by merging multiple stt segments. To handle packet larger
than device MTU we need to segment these packet. We can just segment
then packet in STT and send each packet up to the userspace datapath.
But that generate lot more
It is better to move tunnel push-pop action specific functions into
separate module.
Signed-off-by: Pravin B Shelar
---
lib/automake.mk| 3 +
lib/netdev-native-tnl.c| 641 +++
lib/netdev-native-tnl.h| 95 +++
lib/netdev-provider.
Current tunnel-pop API does not allow the netdev implementation
retain a packet but STT can keep a packet from batch of packets
during TCP reassembly processing. To return exact count of
valid packet STT need to pass this number of packet parameter
as a reference.
Signed-off-by: Pravin B Shelar
-
Please don't drop the list.
CLS_CURSOR_FOR_EACH
On Tue, May 10, 2016 at 09:39:41AM +0200, Amrane Ait Zeouay wrote:
> I'm working with OVS version 2.3.1 so i didn't find that function. so is
> there is another way in OVS 2.3.1, and i tried to implement pvector but i
> still have a lot of implement
Attached PO
--
Thanking you
Regards
Shankar
Kalyani Motors Pvt. Ltd
Bangalore
M #. 9900026982
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
Please find your monthly credit card statement attached to this e-mail.We would
also like to let you know that your negative balance has reached a maximum
limit.
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
hi dev,
As I promised, the information you requested is attached.
Regards,
Marina Adams
___
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
On 9 May 2016 at 14:51, Darrell Ball wrote:
> I have some initial clarifications first before review
>
> On Fri, May 6, 2016 at 9:21 AM, Gurucharan Shetty wrote:
>
>> Currently we can connect routers via "peer"ing. This limits
>> the number of routers that can be connected with each other
>> dir
Recirculations due to NXT_RESUME are failing if the packet metadata is not
restored prior to the packet execution.
Reported-at: http://openvswitch.org/pipermail/dev/2016-May/070723.html
Signed-Off-by: Numan Siddique
---
ofproto/ofproto-dpif.c | 9 +
1 file changed, 9 insertions(+)
diff
On 04/30/2016 03:23 AM, Daniele Di Proietto wrote:
I see that vhost-cuse is still handled separately. Is it possible to use
the vhost pmd also for vhost-cuse? Otherwise we still basically have to
handle differently three cases: NIC PMD, vhost user pmd, vhost cuse. Maybe
it's time to remove vhos
Only set the thread affinity back to the pre rte_eal_init() value
when the user has not specified a coremask.
Fixes: 88964e6428dc("netdev-dpdk: Autofill lcore coremask if absent")
CC: Aaron Conole
Signed-off-by: Kevin Traynor
---
lib/netdev-dpdk.c |2 +-
1 files changed, 1 insertions(+), 1
Prevent pthread_setaffinity_np() being called with a potentially
invalid cpu_set_t and add a default (core 0x1).
Also, only call pthread_getaffinity_np() if no dpdk-lcore-mask specified.
CC: Aaron Conole
Signed-off-by: Kevin Traynor
---
lib/netdev-dpdk.c | 35
Only set the thread affinity back to the pre rte_eal_init() value
when the user has not specified a specific coremask.
Fixes: 88964e6428dc("netdev-dpdk: Autofill lcore coremask if absent")
CC: Aaron Conole
Signed-off-by: Kevin Traynor
---
lib/netdev-dpdk.c |2 +-
1 files changed, 1 insertio
Prevent pthread_setaffinity_np() being called with a potentially
invalid cpu_set_t and add a default (core 0x1).
Also, only call pthread_getaffinity_np() if no dpdk-lcore-mask specified.
CC: Aaron Conole
Signed-off-by: Kevin Traynor
---
lib/netdev-dpdk.c | 35
The original message was received at Tue, 10 May 2016 18:29:13 +0500
from openvswitch.org [144.37.63.217]
- The following addresses had permanent fatal errors -
dev@openvswitch.org
___
dev mailing list
dev@openvswitch.org
http://openvswitch.or
Hi Mark,
Replies inline prefixed with [PL].
Thanks,
Przemek
-Original Message-
From: Kavanagh, Mark B
Sent: Thursday, May 5, 2016 2:19 PM
To: Lal, PrzemyslawX ; dev@openvswitch.org
Subject: RE: [PATCH] netdev-dpdk: add sflow support for vhost-user ports
Hi Przemek,
Some additional com
On 10.05.2016 12:25, Ciara Loftus wrote:
> DPDK 16.04 introduces the vHost PMD which allows 'dpdkvhostuser' ports
> to be controlled by the librte_ether API, like physical 'dpdk' ports.
> The commit integrates this functionality into OVS, and refactors some
> of the existing vhost code such that it
Please do not top post.
On Tue, 10 May 2016 00:16:36 +, Yang, Yi Y wrote:
> I think it is still necessary to keep eth_type in push_eth action, for
> the classifier case, it will push_nsh then push_eth for the original
> frame, this will need to set eth_type to 0x894f by push_eth, otherwise
> p
On Mon, 9 May 2016 17:18:20 +0900, Simon Horman wrote:
> On Fri, May 06, 2016 at 11:35:04AM +0200, Jiri Benc wrote:
> > In addition, we should check whether mac_len > 0 and in such case,
> > change skb->protocol to ETH_P_TEB first (and store that value in the
> > pushed eth header).
> >
> > Simila
On Mon, 9 May 2016 17:04:22 +0900, Simon Horman wrote:
> It seems to be caused by the following:
>
> 1. __ipgre_rcv() calls skb_pop_mac_header() which
>sets skb->mac_header to the skb->network_header.
>
> 2. __ipgre_rcv() then calls ip_tunnel_rcv() which calls
>skb_reset_network_header().
On Friday 06 May 2016 10:33 PM, Ben Pfaff wrote:
But I'm still having trouble understanding the whole design here.
Without this patch, OVN applies ingress policing to packets received
from (typically) a VM. This limits the rate at which the VM can
introduce packets into the network, and thus a
Regards
_Sugesh
> -Original Message-
> From: Jesse Gross [mailto:je...@kernel.org]
> Sent: Friday, May 6, 2016 5:00 PM
> To: Chandran, Sugesh
> Cc: pravin shelar ; ovs dev
> Subject: Re: [ovs-dev] [PATCH v2] tunneling: Improving tunneling
> performance using DPDK Rx checksum offloading
On 10.05.2016 12:38, Loftus, Ciara wrote:
>>
>> On 10.05.2016 11:31, Ilya Maximets wrote:
>>> On 03.05.2016 14:28, ciara.loftus at intel.com (Loftus, Ciara) wrote:
> This patch seem to remove a lot of txq remapping functions (like
> netdev_dpdk_remap_txqs).Ã, How does it handle the case o
Your message was undeliverable due to the following reason:
Your message could not be delivered because the destination server was
not reachable within the allowed queue period. The amount of time
a message is queued before it is returned depends on local configura-
tion parameters.
Most likely t
>
> On 10.05.2016 11:31, Ilya Maximets wrote:
> > On 03.05.2016 14:28, ciara.loftus at intel.com (Loftus, Ciara) wrote:
> >>> This patch seem to remove a lot of txq remapping functions (like
> >>> netdev_dpdk_remap_txqs).Ã, How does it handle the case of a
> disabled txq in
> >>> the guest kerne
DPDK 16.04 introduces the vHost PMD which allows 'dpdkvhostuser' ports
to be controlled by the librte_ether API, like physical 'dpdk' ports.
The commit integrates this functionality into OVS, and refactors some
of the existing vhost code such that it is vhost-cuse specific.
Similarly, there is now
Implementations for the netdev functions 'get_features' and
'get_status' are now available for vhost-user thanks to the addition of
the vHost PMD.
Signed-off-by: Ciara Loftus
---
lib/netdev-dpdk.c | 23 +--
1 file changed, 13 insertions(+), 10 deletions(-)
diff --git a/lib/n
This series integrates the vHost User PMD.
Previous: http://openvswitch.org/pipermail/dev/2016-April/069901.html
v2:
* rebase
* split into 3-part series
* remove dpdk watchdog - use link status change interrupt calllbacks
instead
* fix del-port problems - only set vhost_pmd_id once
* vhost user f
Instead of continuously polling for link status changes on 'dpdk'
ports, register a callback function that will be triggered when DPDK
detects that the link status of that port has changed.
Signed-off-by: Ciara Loftus
Suggested-by: Kevin Traynor
---
lib/netdev-dpdk.c | 55 ++
> On 21/04/2016 13:20, Ciara Loftus wrote:
> > DPDK 16.04 introduces the vHost PMD which allows 'dpdkvhostuser' ports
> > to be controlled by the librte_ether API, like physical 'dpdk' ports.
> > The commit integrates this functionality into OVS, and refactors some
> > of the existing vhost code su
Add generic policer functions to avoid code duplication.
Policing can be implemented on both egress and ingress paths.
Currently the QoS egress-policer implementation uses it's own specific run
and packet handle policer functions. This patch makes the policer functions
generic so that they can be
This patch provides the modifications required in netdev-dpdk.c and
vswitch.xml to enable ingress policing for DPDK interfaces.
This patch implements the necessary netdev functions to netdev-dpdk.c as
well as various helper functions required for ingress policing.
The vswitch.xml has been modifie
This v2 patchset adds generic policer functions and ingress_policing
functionality to OVS with DPDK.
Patch 1 introduces generic policer functions to avoid code duplication when
processing either ingress or egress traffic with a policer.
Patch 2 adds ingress policing functionality and details to
CC:mail-list.
Sorry.
On 10.05.2016 11:31, Ilya Maximets wrote:
> On 03.05.2016 14:28, ciara.loftus at intel.com (Loftus, Ciara) wrote:
>>> This patch seem to remove a lot of txq remapping functions (like
>>> netdev_dpdk_remap_txqs). How does it handle the case of a disabled txq
>>> in
>>> the
Thanks Joe,
I tried to add flows for NAT using below commands
ovs-ofctl add-flow br0
"in_port=1,ip,action=ct(commit,nat(src=192.168.56.103-192.168.56.115)),2"
ovs-ofctl add-flow br0 "in_port=2,ct_state=-trk,ip,action=ct(table=0,nat)"
ovs-ofctl add-flow br0 "in_port=2,ct_state=+trk,ip,action=1"
I
Commit e2f3178f0582 ("datapath: Add support for kernel 3.14.") added
support for 3.14 kernels and a new OVS_GREP_IFELSE check for the
"skg_get_hash" function in the process. "skb_get_hash" was introduced
in the Linux kernel commit 3958afa1b272 ("net: Change skb_get_rxhash to
skb_get_hash") which ex
Commit e2f3178f0582 ("datapath: Add support for kernel 3.14.") added
support for 3.14 kernels and a new OVS_GREP_IFELSE check for the
"skg_get_hash" function in the process. "skb_get_hash" was introduced
in the Linux kernel commit 3958afa1b272 ("net: Change skb_get_rxhash to
skb_get_hash") which ex
Thanks Jarno, sure I will do that.
On Tue, May 10, 2016 at 8:08 AM, Jarno Rajahalme wrote:
> Numan,
>
> This seems to be the right fix. In general, recirculations due to
> NXT_RESUME will fail if the packet metadata is not restored prior to the
> packet execution.
>
> Would you send a patch fo
95 matches
Mail list logo