During E-W routing for vlan backed networks, we replace router port
mac with chassis mac, when packet leaves the source hypervisor.
As a result, the destination VM (on remote hypervisor) will see
chassis mac as source mac in received packet.
Although, functionality wise this does not cause any
Hi Numan,
Thanks a lot for the feedback.
Sure, I can define the range for physical flows.
However, just wanted to call out that as I understand, conjunction id has to
be unique only across a table,
i.e IDs can overlap across different openflow tables (physical table and
logical flow tables
Seems some issue in my local env where --enable-Werror --enable-sparse
didn't catch this warning. Addressed in v4.
On Mon, Aug 26, 2019 at 3:01 PM 0-day Robot wrote:
> Bleep bloop. Greetings aginwala aginwala, I am a robot and I have tried
> out your patch.
> Thanks for your contribution.
>
>
From: Aliasgar Ginwala
Add support in ovsdb-tool for migrating clustered dbs to standalone dbs.
E.g. usage to migrate nb/sb db to standalone db from raft:
ovsdb-tool cluster-to-standalone ovnnb_db.db ovnnb_db_cluster.db
Signed-off-by: Aliasgar Ginwala
---
Documentation/ref/ovsdb.7.rst | 3 +
The patch adds support for using need_wakeup flag in AF_XDP rings.
When this flag is used, it means that OVS has to explicitly wake
up the kernel RX, using poll() syscall and wake up TX, using sendto()
syscall. This feature improves the performance by avoiding unnecessary
syscalls, so keeping more
Make the spin lock struct 64-byte aligned to avoid false sharing issue.
Signed-off-by: William Tu
---
include/openvswitch/thread.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/openvswitch/thread.h b/include/openvswitch/thread.h
index acc822904847..5053cb3091c3
Bleep bloop. Greetings aginwala aginwala, I am a robot and I have tried out
your patch.
Thanks for your contribution.
I encountered some error that I wasn't expecting. See the details below.
build:
/bin/sh ./libtool --tag=CC --mode=link gcc -std=gnu99 -Wstrict-prototypes
-Wall -Wextra
Hi Han and Ben:
Addressed review comments from both of you in v3
https://patchwork.ozlabs.org/patch/1153456/ . PTAL
On Fri, Aug 23, 2019 at 3:37 PM Ginwala, Aliasgar wrote:
> Thanks Ben:
>
> On 8/23/19, 2:50 PM, "Ben Pfaff" wrote:
>
> On Fri, Aug 23, 2019 at 09:29:00PM +, Ginwala,
From: Aliasgar Ginwala
Add support in ovsdb-tool for migrating clustered dbs to standalone dbs.
E.g. usage to migrate nb/sb db to standalone db from raft:
ovsdb-tool cluster-to-standalone ovnnb_db.db ovnnb_db_cluster.db
Signed-off-by: Aliasgar Ginwala
---
Documentation/ref/ovsdb.7.rst | 3 +
Requested-by: Leonid Ryzhyk
Signed-off-by: Ben Pfaff
---
Documentation/ref/ovsdb-server.7.rst | 9 +
NEWS | 3 ++-
ovsdb/execution.c| 26 ++
ovsdb/transaction.c | 22 +-
Please don't send HTML-only email. I generally ignore it.
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
On Fri, Aug 23, 2019 at 9:09 AM Darrell Ball wrote:
> Thanks for the patch
>
> Goes back to release 2.6/day one :-).
>
> I'll provide more feedback after today.
>
I sent an alternative patch here
https://mail.openvswitch.org/pipermail/ovs-dev/2019-August/362013.html
pls take a look.
>
> On
On Thu, Aug 8, 2019 at 10:57 AM Vishal Deep Ajmera
wrote:
>
> Problem:
>
> In OVS-DPDK, flows with output over a bond interface of type “balance-tcp”
> (using a hash on TCP/UDP 5-tuple) get translated by the ofproto layer into
> "HASH" and "RECIRC" datapath actions. After recirculation,
Resent this patch, as it had a bad e-mail address
Darrell
On Mon, Aug 26, 2019 at 9:06 AM Darrell Ball wrote:
> The ICMP error data L4 length check was found to be too strict for TCP,
> expecting a minimum of 20 rather than 8 bytes. This worked by
> hapenstance for other inner protocols. The
The ICMP error data L4 length check was found to be too strict for TCP,
expecting a minimum of 20 rather than 8 bytes. This worked by
hapenstance for other inner protocols. The approach is to explicitly
handle the ICMPV4 error data L4 length check and to do this for all
supported inner protocols
On 8/26/2019 3:12 PM, Ilya Maximets wrote:
On 23.07.2019 14:20, Ilya Maximets wrote:
If the port was destroyed during the initial reconfiguration, we should
report an error to the upper layers. Otherwise, successful addition of
the port will be logged and upper layers will continue to
The ICMP error data L4 length check was found to be too strict for TCP,
expecting a minimum of 20 rather than 8 bytes. This worked by
hapenstance for other inner protocols. The approach is to explicitly
handle the ICMPV4 error data L4 length check and to do this for all
supported inner protocols
This may be needed in some special cases, such as to support some hardware
offload implementations. Note that disabling TCP sequence number
verification is not an optimization in itself, but supporting some
hardware offload implementations may offer better performance. TCP
sequence number
Cache the 'conn' context and use it when it is valid. The cached 'conn'
context will get reset if it is not expected to be valid; the cost to do
this is negligible. Besides being most optimal, this also handles corner
cases, such as decapsulation leading to the same tuple, as in tunnel VPN
'tx_q' array is allocated for each DPDK netdev. 'struct dpdk_tx_queue'
is 8 bytes long, so 8 tx queues are sharing the same cache line in
case of 64B cacheline size. This causes 'false sharing' issue in
mutliqueue case because taking the spinlock implies write to memory
i.e. cache invalidation.
Add a coverage counter to help diagnose contention on the vhost txqs.
This is seen as dropped packets on the physical ports for rates that
are usually handled fine by OVS.
Signed-off-by: David Marchand
---
lib/netdev-dpdk.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff
On Mon, Jul 8, 2019 at 4:14 PM Ilya Maximets wrote:
>
> On 07.07.2019 19:13, David Marchand wrote:
> > Add a statistic to help diagnose contention on the vhost txqs.
> > This is usually seen as dropped packets on the physical ports for rates
> > that are usually handled fine by OVS.
> >
> >
On 23.07.2019 14:20, Ilya Maximets wrote:
> If the port was destroyed during the initial reconfiguration, we should
> report an error to the upper layers. Otherwise, successful addition of
> the port will be logged and upper layers will continue to configure
> this port. For example, the 'dpif'
On 19.08.2019 14:18, Ilya Maximets wrote:
> Version 2:
> * Fixed incorrect handling of vhost stats if there are more than
> one of them.
>
> Ilya Maximets (2):
> netdev-dpdk: Fix not reporting rx_oversize_errors in stats.
> netdev-dpdk: Refactor vhost custom stats for extensibility.
>
Bleep bloop. Greetings Sriram Vatala via dev, I am a robot and I have tried
out your patch.
Thanks for your contribution.
I encountered some error that I wasn't expecting. See the details below.
checkpatch:
WARNING: Line is 80 characters long (recommended limit is 79)
#247 FILE:
Hi.
Not a full review. Few comments inline.
Best regards, Ilya Maximets.
On 08.08.2019 11:56, Vishal Deep Ajmera wrote:
> Problem:
>
> In OVS-DPDK, flows with output over a bond interface of type “balance-tcp”
> (using a hash on TCP/UDP 5-tuple) get translated by the ofproto layer into
OVS may be unable to transmit packets for multiple reasons and
today there is a single counter to track packets dropped due to
any of those reasons. The most common reason is that a VM is
unable to read packets fast enough causing the vhostuser port
transmit queue on the OVS side to become full.
Hi,
has any one see the following backtrace?
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
Core was generated by `ovs-vswitchd unix:/var/run/openvswitch/db.sock
-vconsole:emer -vsyslog:err -vfi'.
Program terminated with signal SIGABRT, Aborted.
#0 __GI_raise
Hi Ankur,
This patch doesn't apply now. Can you please re base. Few comments below
Thanks
Numan
On Thu, Aug 22, 2019 at 1:51 AM Ankur Sharma
wrote:
> During E-W routing for vlan backed networks, we replace router port
> mac with chassis mac, when packet leaves the source hypervisor.
>
> As a
On Sun, Aug 25, 2019 at 11:32 PM Lorenzo Bianconi <
lorenzo.bianc...@redhat.com> wrote:
> Fix memory leak of ip_address string in build_pre_lb routine if we
> install logical flows for empty_lb controller event
>
> Fixes: f49b17a6cbe3 ("OVN: use trigger_event action to report
> 'empty_lb_rule'
Hi Darrell,
Can we get this merged into master as well as backports to affected branches?
Warm Regards,
Vishal Ajmera
___
dev mailing list
d...@openvswitch.org
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
Fix memory leak of ip_address string in build_pre_lb routine if we
install logical flows for empty_lb controller event
Fixes: f49b17a6cbe3 ("OVN: use trigger_event action to report 'empty_lb_rule'
events")
Signed-off-by: Lorenzo Bianconi
---
ovn/northd/ovn-northd.c | 1 -
1 file changed, 1
On Tue, Aug 13, 2019 at 6:35 PM Flavio Leitner via dev
wrote:
>
> When a packet needs to be encapsulated in userspace, the endpoint
> address needs to be resolved to fill in the headers. If it is not,
> then currently OvS sends either a Neighbor Solicitation (IPv6)
> or an ARP Query (IPv4) to
33 matches
Mail list logo