> On Jun 27, 2017, at 12:39 PM, Ben Pfaff wrote:
>
> On Sat, Jun 24, 2017 at 10:10:02AM +0800, Justin Pettit wrote:
>>
>>> On Jun 24, 2017, at 2:12 AM, Joe Stringer wrote:
>>>
On 23 June 2017 at 05:38, Russell Bryant wrote:
> On Thu, Jun
This function is a wrapper for parse_ofp_packet_out_str__(), which
states that a couple members must be freed on success. This commit
updates the wrapper's description to indicate the same.
The existing callers appear to be properly freeing members.
Signed-off-by: Justin Pettit
The function parse_ofp_meter_mod_str() allocates a buffer called
'bands', which parse_ofp_meter_mod_str__() then steals for the member
'mm->meter.bands'. Calling functions didn't free that stolen value and
the comments for those function didn't indicate that was necessary.
Found by valgrind.
This patch has an issue. I'm working on a v2.
--Justin
> On Jun 27, 2017, at 5:35 PM, Justin Pettit wrote:
>
> The function parse_ofp_meter_mod_str() allocates a buffer called
> 'bands', which parse_ofp_meter_mod_str__() then steals for the member
> 'mm->meter.bands'.
The function parse_ofp_meter_mod_str() allocates a buffer called
'bands', which parse_ofp_meter_mod_str__() then steals for the member
'mm->meter.bands'. Calling functions didn't free that stolen value and
the comments for those function didn't indicate that was necessary.
Found by valgrind.
This function is a wrapper for parse_ofp_packet_out_str__(), which
states that a couple members must be freed on success. This commit
updates the wrapper's description to indicate the same.
The existing callers appear to be properly freeing members.
Signed-off-by: Justin Pettit
On 06/26/2017 09:05 PM, 王志克 wrote:
Hi Joe, Greg,
I tried to create a pull request, please check whether it works. Thanks.
https://github.com/openvswitch/ovs/pull/187
I've been trying to test this on a Centos 7.2 3.10 kernel but am
running into unrelated build issues. I'll see if I can get a
En línea y en Vivo / Para todo su Equipo con una sola Conexión
Cómo Proteger la Información Confidencial, Secretos Industriales y Comerciales
Su Blindaje y Protección Legal
19 de julio - Online en Vivo - 10:00 a 13:00 y de 15:00 a 18:00 Hrs
El objetivo de esta Capacitación Online en Vivo
On 27 June 2017 at 13:40, Darrell Ball wrote:
>
>
> On 6/27/17, 1:20 PM, "Joe Stringer" wrote:
>
> On 27 June 2017 at 11:19, Darrell Ball wrote:
> >
> >
> > On 6/27/17, 11:04 AM, "Joe Stringer" wrote:
> >
>
On Fri, Jun 23, 2017 at 09:17:01AM -0700, Greg Rose wrote:
> On 06/19/2017 04:30 PM, Ben Pfaff wrote:
> >This wasn't clear before.
> >
> >Reported-at:
> >https://mail.openvswitch.org/pipermail/ovs-dev/2017-June/334271.html
> >Signed-off-by: Ben Pfaff
> >---
> >
On Fri, Jun 23, 2017 at 04:47:48PM +, Zoltán Balogh wrote:
>
> This series was started by Ben Pfaff, v3 can be found here:
> https://patchwork.ozlabs.org/patch/778070/
> https://patchwork.ozlabs.org/patch/778071/
> https://patchwork.ozlabs.org/patch/778076/
>
On 6/27/17, 1:20 PM, "Joe Stringer" wrote:
On 27 June 2017 at 11:19, Darrell Ball wrote:
>
>
> On 6/27/17, 11:04 AM, "Joe Stringer" wrote:
>
> On 27 June 2017 at 10:51, Darrell Ball wrote:
>
>On 26.06.2017 00:52, Bodireddy, Bhanuprakash wrote:
+
+/* Flush the txq if there are any packets available.
+ * dynamic_txqs/concurrent_txq is disabled for vHost User ports as
+ * 'OVS_VHOST_MAX_QUEUE_NUM' txqs are preallocated.
+ */
>>>
>>> This comment is completely
On 27 June 2017 at 11:19, Darrell Ball wrote:
>
>
> On 6/27/17, 11:04 AM, "Joe Stringer" wrote:
>
> On 27 June 2017 at 10:51, Darrell Ball wrote:
> >
> >
> > On 6/27/17, 10:47 AM, "Joe Stringer" wrote:
> >
>
On 27 June 2017 at 12:48, Anand Kumar wrote:
> Hi Guru,
>
> Could you please apply this patch?
>
Applied, thanks for the reminder.
>
> Thanks,
> Anand Kumar
>
> On 6/23/17, 3:04 PM, "Sairam Venugopal" wrote:
>
> Acked-by: Sairam Venugopal
On Sat, Jun 24, 2017 at 10:10:02AM +0800, Justin Pettit wrote:
>
> > On Jun 24, 2017, at 2:12 AM, Joe Stringer wrote:
> >
> >> On 23 June 2017 at 05:38, Russell Bryant wrote:
> >>> On Thu, Jun 8, 2017 at 12:12 PM, Ben Pfaff wrote:
> On Thu,
Hi Guru,
Could you please apply this patch?
Thanks,
Anand Kumar
On 6/23/17, 3:04 PM, "Sairam Venugopal" wrote:
Acked-by: Sairam Venugopal
On 6/23/17, 2:05 PM, "ovs-dev-boun...@openvswitch.org on behalf of Anand
Kumar"
On 6/27/17, 11:04 AM, "Joe Stringer" wrote:
On 27 June 2017 at 10:51, Darrell Ball wrote:
>
>
> On 6/27/17, 10:47 AM, "Joe Stringer" wrote:
>
> On 27 June 2017 at 10:42, Darrell Ball wrote:
>
Commit 32b77c316d9982("dpif: Save added ports in a port map.")
introduced tracking of all dpif ports by taking a reference on each
available netdev when the dpif is opened, but it failed to clear out and
release references to these netdevs when the dpif is closed.
Balance the referencing of
An upcoming patch will reuse this from elsewhere.
Signed-off-by: Joe Stringer
---
lib/netdev.c | 14 ++
1 file changed, 10 insertions(+), 4 deletions(-)
diff --git a/lib/netdev.c b/lib/netdev.c
index 765bf4b9ccad..eb7aef7376f1 100644
--- a/lib/netdev.c
+++
Previously, a user need to run ofproto/trace multiple times to derive the
final datapath actions if a flow hit conntrack actions that involves
recirculation. To improve the usability of ofproto/trace, in this patch,
we keep track of the conntrack actions, and automatically run the
recirculation
This patch series enhances the usability of ofproto/trace when a flow
hit conntrack actions that involves recirculation. Instead of asking
users to run ofproto/trace multiple times to derive the final datapath
actions, we keep track of the conntrack actions, and automatically run
the recirculation
On 26 June 2017 at 19:42, Darrell Ball wrote:
> Hardware offload introduced extra tracking of netdev ports. This
> included ovs-netdev, which is really for internal infra usage for
> the userpace datapath. This breaks cleanup of the userspace
> datapath. One effect is that
On 27 June 2017 at 10:51, Darrell Ball wrote:
>
>
> On 6/27/17, 10:47 AM, "Joe Stringer" wrote:
>
> On 27 June 2017 at 10:42, Darrell Ball wrote:
> >
> >
> > On 6/27/17, 10:40 AM, "Joe Stringer" wrote:
> >
>
On 27 June 2017 at 08:12, Chandran, Sugesh wrote:
> Hi Joe,
> Thank you for looking into the patch series.
> Please find my answers inline below.
>
>
>
> Regards
> _Sugesh
>
>
>> -Original Message-
>> From: Joe Stringer [mailto:j...@ovn.org]
>> Sent: Monday,
One more suggestion:
Change
While pmd threads are created based on pmd-cpu-mask, the
thread only starts consuming CPU cycles if there is least one receive queue
assigned to the pmd.
to
Even though a PMD thread may exist, the thread only starts consuming CPU
cycles if there is least one
On 6/27/17, 10:47 AM, "Joe Stringer" wrote:
On 27 June 2017 at 10:42, Darrell Ball wrote:
>
>
> On 6/27/17, 10:40 AM, "Joe Stringer" wrote:
>
> On 26 June 2017 at 18:19, Darrell Ball wrote:
>
On 27 June 2017 at 10:42, Darrell Ball wrote:
>
>
> On 6/27/17, 10:40 AM, "Joe Stringer" wrote:
>
> On 26 June 2017 at 18:19, Darrell Ball wrote:
> >
> >
> > On 6/26/17, 4:49 PM, "Joe Stringer" wrote:
> >
>
On Tue, Jun 27, 2017 at 10:40 AM, Han Zhou wrote:
>
>
>
> On Tue, Jun 27, 2017 at 10:12 AM, Mickey Spiegel
wrote:
> >
> >
> > On Tue, Jun 27, 2017 at 1:02 AM, Han Zhou wrote:
> >>
> >> Localnet port was supposed to work for directly
On 6/27/17, 10:40 AM, "Joe Stringer" wrote:
On 26 June 2017 at 18:19, Darrell Ball wrote:
>
>
> On 6/26/17, 4:49 PM, "Joe Stringer" wrote:
>
> On 26 June 2017 at 13:22, Joe Stringer wrote:
> > On
On Tue, Jun 27, 2017 at 10:12 AM, Mickey Spiegel
wrote:
>
>
> On Tue, Jun 27, 2017 at 1:02 AM, Han Zhou wrote:
>>
>> Localnet port was supposed to work for directly connected datapath
>> only. However, the recursive local_datapath filling introduced a
>>
On 26 June 2017 at 18:19, Darrell Ball wrote:
>
>
> On 6/26/17, 4:49 PM, "Joe Stringer" wrote:
>
> On 26 June 2017 at 13:22, Joe Stringer wrote:
> > On 26 June 2017 at 11:45, Darrell Ball wrote:
> >> On 6/26/17, 10:22
Let us cover the documentation first
On 6/27/17, 8:46 AM, "ovs-dev-boun...@openvswitch.org on behalf of Billy
O'Mahony" wrote:
Previously if there is no available (non-isolated) pmd on the numa node
for a port then
On 27 June 2017 at 09:30, Chandran, Sugesh wrote:
>
>
> Regards
> _Sugesh
>
>
>> -Original Message-
>> From: Joe Stringer [mailto:j...@ovn.org]
>> Sent: Monday, June 26, 2017 7:06 PM
>> To: Chandran, Sugesh
>> Cc: ovs dev
On Tue, Jun 27, 2017 at 1:02 AM, Han Zhou wrote:
> Localnet port was supposed to work for directly connected datapath
> only. However, the recursive local_datapath filling introduced a
> problem in below scenario:
>
> LS A <-> LR <-> LS B, port a@HV1 is on LS A, port b@HV2 is
It is not about limit but more about use case. Could you explain your use
case why using localnet ports here while the VMs can be connected through
logical router on overlay? Did my patch work for you? Does it work also
when you remove the localnet ports?
Mickey mentioned scenarios when a
On Thu, Jun 15, 2017 at 1:04 AM, wrote:
> Hi Russell, I am sorry for the late reply.
> The route not bound to a chassis, and have no redirect-chassis. The dumped
> northbound db is as follow.
> Ip addresses of 100.0.0.148 and 200.0.0.2 locate on different chassis. The
>
Regards
_Sugesh
> -Original Message-
> From: Joe Stringer [mailto:j...@ovn.org]
> Sent: Monday, June 26, 2017 7:06 PM
> To: Chandran, Sugesh
> Cc: ovs dev ; Zoltán Balogh
> ; Andy Zhou
>
I'll give Darrell a chance to comment before rev'ing.
> -Original Message-
> From: Ilya Maximets [mailto:i.maxim...@samsung.com]
> Sent: Tuesday, June 27, 2017 5:11 PM
> To: O Mahony, Billy ; d...@openvswitch.org
> Cc: dlu...@gmail.com
> Subject: Re: [PATCH v7]
Previously if there is no available (non-isolated) pmd on the numa node
for a port then the port is not polled at all. This can result in a
non-operational system until such time as nics are physically
repositioned. It is preferable to operate with a pmd on the 'wrong' numa
node albeit with lower
I have incorporated Darrell's doc comments and Ilya's implementation and
tested with various scenarios. I had to make a few small changes to ensure that
the warning messages were issued for each queue as per previous patch comments.
Billy O'Mahony (1):
dpif-netdev: Assign ports to pmds on
Regards
_Sugesh
> -Original Message-
> From: Joe Stringer [mailto:j...@ovn.org]
> Sent: Monday, June 26, 2017 6:30 PM
> To: Chandran, Sugesh
> Cc: ovs dev ; Zoltán Balogh
> ; Andy Zhou
>
Regards
_Sugesh
> -Original Message-
> From: Joe Stringer [mailto:j...@ovn.org]
> Sent: Monday, June 26, 2017 6:27 PM
> To: Chandran, Sugesh
> Cc: ovs dev ; Zoltán Balogh
>
> Subject: Re: [PATCH 2/4]
Hi Joe,
Thank you for looking into the patch series.
Please find my answers inline below.
Regards
_Sugesh
> -Original Message-
> From: Joe Stringer [mailto:j...@ovn.org]
> Sent: Monday, June 26, 2017 6:25 PM
> To: Chandran, Sugesh
> Cc: ovs dev
On 06/26/2017 05:51 PM, 王志克 wrote:
Hi Greg,
The exact issue occured on the 20th of check-kmod (sometimes there are other
kernel issue: kernel just hangs but without panic). OVS2.6.0 on CentOS7.2 with
kernel 3.10.0-327.el7.x86_64. Some info below, which hopes helpful.
OK, I'll try with that
Avoiding checksum validation in conntrack module if it is already verified
in DPDK physical NIC ports.
Signed-off-by: Sugesh Chandran
---
lib/conntrack.c | 71 ++---
lib/dp-packet.h | 22 ++
2 files
DPDK ports use masks while reporting rx checksum flags. OVS should use these
mask along with reported checksum flag while validating the rx checksum.
Signed-off-by: Sugesh Chandran
Suggested-by: Darrell Ball
---
lib/dp-packet.h | 6 --
1 file
Conntrack need not verify the checksum of incoming packets if its validated by
DPDK physical NIC ports.
Also make use the DPDK rx checksum mask bits along with flags while validating
the reported hardware checksum state.
v2->v3
- Rebased on latest master.
- Updated the existing checksum
Regards
_Sugesh
> -Original Message-
> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> boun...@openvswitch.org] On Behalf Of Chandran, Sugesh
> Sent: Tuesday, June 27, 2017 11:48 AM
> To: Darrell Ball
> Cc: d...@openvswitch.org
> Subject: Re: [ovs-dev]
Hi Darrel,
Regards
_Sugesh
> -Original Message-
> From: Darrell Ball [mailto:db...@vmware.com]
> Sent: Tuesday, June 27, 2017 8:15 AM
> To: Chandran, Sugesh
> Cc: d...@openvswitch.org
> Subject: Re: [PATCH v2] conntrack : Use Rx checksum offload feature on
>
Patch is marked as RFC because DPDK 17.05.1 is not yet
published.
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:
Hi Han Zhou,
> If using localnet, it should rely on physical network (L2 and L3) to
reach the destination, not overlay, so adding the logical router here
doesn't make sense here
Why ovn have this limit for physical network? Does this mean that vlan
network can not use the l3 function of ovn?
On Thu, Jun 15, 2017 at 1:04 AM, wrote:
>
> Hi Russell, I am sorry for the late reply.
> The route not bound to a chassis, and have no redirect-chassis. The dumped
> northbound db is as follow.
> Ip addresses of 100.0.0.148 and 200.0.0.2 locate on different chassis. The
>
Localnet port was supposed to work for directly connected datapath
only. However, the recursive local_datapath filling introduced a
problem in below scenario:
LS A <-> LR <-> LS B, port a@HV1 is on LS A, port b@HV2 is on LS B.
If B has localnet port, then ovn-controller on HV1 would think port
b
Hi Sugesh
On 6/26/17, 12:51 AM, "Chandran, Sugesh" wrote:
Hi Darrel,
Please find my answers below.
Regards
_Sugesh
> -Original Message-
> From: Darrell Ball [mailto:db...@vmware.com]
> Sent: Saturday, June 24, 2017 10:39
tart a virtual machine with its backend tap device attached to a brought up
linux bridge.
If we delete the linux bridge when vm is still running, we'll get the following
error when
trying to create a ovs bridge with the same name.
The reason is that ovs-router subsystem add the linux bridge
56 matches
Mail list logo