Hi Daniele,
I just started to have a look to your new v2 patch set.
A minor comment - I know this is a bit of nit-picking - but I ran
utilities/checkpatch.py and I got some output like
patch 4/15
W(1692): Line has trailing whitespace
+static inline void ct_lock_lock(struct ct_lock *lock)
patch 5/
hanks,
Antonio
> -Original Message-
> From: Ben Pfaff [mailto:b...@ovn.org]
> Sent: Thursday, April 21, 2016 7:41 PM
> To: Fischetti, Antonio
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH RFC] dpif-netdev: ACL+dpcls for Wildcard
> matching.
>
> On
Hi Daniele,
below are
- Regression Tests results
- ALWAYS_INLINE check
Regression Tests on Performance
===
I did some regression tests on performance, it looks like there's
no performance drop.
I've sent mono-directional data with 64-Byte packets at line rate
o
Hi Daniele,
some comments inline prefixed with [Antonio F].
Regards,
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele Di
> Proietto
> Sent: Saturday, April 16, 2016 1:03 AM
> To: dev@openvswitch.org
> Subject: [ovs-dev] [PATCH v2 04/15] c
Hi Daniele, few comments inline.
> -Original Message-
> From: Daniele Di Proietto [mailto:diproiet...@vmware.com]
> Sent: Wednesday, April 27, 2016 7:36 AM
> To: Fischetti, Antonio
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v2 04/15] conntrack: New user
ort dest, ToS, VlanTci.
=> dpcls used 6 subtables, max throughtput went down to: 2.2 Mpps
> -Original Message-
> From: Ben Pfaff [mailto:b...@ovn.org]
> Sent: Monday, May 2, 2016 10:16 PM
> To: Fischetti, Antonio
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATC
Hi Jarno, my reply inline.
Thanks,
Antonio
> -Original Message-
> From: Jarno Rajahalme [mailto:ja...@ovn.org]
> Sent: Monday, May 2, 2016 11:05 PM
> To: Fischetti, Antonio
> Cc: Ben Pfaff ; dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH RFC] dpif-netdev: ACL+dp
Thanks for your feedback, Jarno. Replies inline.
From: Jarno Rajahalme [mailto:ja...@ovn.org]
Sent: Wednesday, May 4, 2016 7:23 PM
To: Fischetti, Antonio
Cc: Ben Pfaff ; dev@openvswitch.org
Subject: Re: [ovs-dev] [PATCH RFC] dpif-netdev: ACL+dpcls for Wildcard matching.
On May 4, 2016, at 3:56
From: Jarno Rajahalme [mailto:ja...@ovn.org]
Sent: Thursday, May 5, 2016 7:45 PM
To: Fischetti, Antonio
Cc: Ben Pfaff ; dev@openvswitch.org
Subject: Re: [ovs-dev] [PATCH RFC] dpif-netdev: ACL+dpcls for Wildcard matching.
On May 5, 2016, at 9:27 AM, Fischetti, Antonio
mailto:antonio.fische
c=5.5.5.$i,action=output:2
done
Thanks,
Antonio
> -Original Message-
> From: Ben Pfaff [mailto:b...@ovn.org]
> Sent: Monday, May 2, 2016 10:16 PM
> To: Fischetti, Antonio
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH RFC] dpif-netdev: ACL+dpcls for
> Wi
Hi Jan, thanks for your feedback, some replies below.
Regards,
Antonio
> -Original Message-
> From: Jan Scheurich [mailto:jan.scheur...@ericsson.com]
> Sent: Thursday, May 19, 2016 3:55 PM
> To: Fischetti, Antonio ; Ben Pfaff
>
> Cc: dev@openvswitch.org
> Subject:
One question below about data on real use-cases,
thanks.
> -Original Message-
> From: Jarno Rajahalme [mailto:ja...@ovn.org]
> Sent: Thursday, May 19, 2016 7:51 PM
> To: Fischetti, Antonio
> Cc: Jan Scheurich ; Ben Pfaff
> ; dev@openvswitch.org
> Subject: Re: [ovs-
From: Jarno Rajahalme [mailto:ja...@ovn.org]
Sent: Friday, May 20, 2016 8:59 PM
To: Fischetti, Antonio
Cc: Jan Scheurich ; Ben Pfaff ;
dev@openvswitch.org
Subject: Re: [ovs-dev] [PATCH RFC] dpif-netdev: ACL+dpcls for Wildcard matching.
On May 20, 2016, at 4:07 AM, Fischetti, Antonio
Hi Daniele,
I added one comment below.
Regards,
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele
> Di Proietto
> Sent: Tuesday, May 17, 2016 1:57 AM
> To: dev@openvswitch.org
> Subject: [ovs-dev] [PATCH v3 05/16] conntrack: Periodically d
Thanks Daniele, you did answer my questions.
Below my comments.
> -Original Message-
> From: Daniele Di Proietto [mailto:diproiet...@vmware.com]
> Sent: Tuesday, May 24, 2016 8:32 AM
> To: Fischetti, Antonio
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v
Hi Daniele, just a comment below.
Apart from that, it looks good to me, thanks.
Acked-by: Antonio Fischetti
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele
> Di Proietto
> Sent: Tuesday, May 17, 2016 1:56 AM
> To: dev@openvswitch.org
> Subject:
Looks ok to me.
Acked-by: Antonio Fischetti
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele
> Di Proietto
> Sent: Tuesday, May 17, 2016 1:56 AM
> To: dev@openvswitch.org
> Subject: [ovs-dev] [PATCH v3 03/16] flow: Introduce parse_dl_type().
>
>
Hi Daniele, one question below.
Thanks,
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele
> Di Proietto
> Sent: Tuesday, May 17, 2016 1:57 AM
> To: dev@openvswitch.org
> Subject: [ovs-dev] [PATCH v3 08/16] dpif-netdev: Execute conntrack
>
> -Original Message-
> From: Joe Stringer [mailto:j...@ovn.org]
> Sent: Tuesday, May 24, 2016 7:26 PM
> To: Fischetti, Antonio
> Cc: Daniele Di Proietto ; dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v3 04/16] conntrack: New userspace
> connection tracker.
&
It looks ok to me.
Thanks.
Acked-by: Antonio Fischetti
> -Original Message-
> From: Daniele Di Proietto [mailto:diproiet...@vmware.com]
> Sent: Wednesday, May 25, 2016 9:57 PM
> To: Fischetti, Antonio
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v3
Hi Daniele,
do you have any performance figures related to this implementation?
I know that performance may change quite a lot depending on the type
of traffic data.
Thanks,
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele
> Di Proietto
>
Hi Robert,
one comment below.
I've checked it applies cleanly to the latest master branch.
Also with utilities/checkpatch.py is ok.
Thanks,
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Robert
> Wojciechowicz
> Sent: Monday, May 23, 2016 11:29
Ok, it makes sense.
It looks ok to me now.
Acked-by: Antonio Fischetti
> -Original Message-
> From: Aaron Conole [mailto:acon...@redhat.com]
> Sent: Tuesday, May 31, 2016 7:20 PM
> To: Fischetti, Antonio
> Cc: Wojciechowicz, RobertX ;
> dev@openvswitch.org
>
Hi Ado,
are you're referring to OVS in userspace?
If that's the case, have a look in lib/dpif-netdev.c.
You can start from dp_netdev_process_rxq_port() which receives
the packets. All functions like emc_*() refer to the Microflows
while those as dpcls_*() are for the Megaflow management.
BTW, for
Thanks Daniele,
Antonio
> -Original Message-
> From: Daniele Di Proietto [mailto:diproiet...@vmware.com]
> Sent: Wednesday, June 1, 2016 7:34 PM
> To: Fischetti, Antonio
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v3 00/16] Userspace (DPDK) connection
Hi Jan,
I added some comments prefixed with [Antonio F].
Thanks,
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Jan
> Scheurich
> Sent: Friday, July 15, 2016 5:35 PM
> To: dev@openvswitch.org
> Subject: [ovs-dev] [PATCH v2] dpif-netdev: dpcls pe
Hi Jan,
I don't know if you're going to split this in different patches to
separate cpvector stuff, in any case all the content looks good
to me.
Thanks.
Acked-by: Antonio Fischetti
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Jan
> Scheurich
> Sen
Hi Jan,
that's an interesting approach.
I added few comments and questions below.
Thanks,
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Jan
> Scheurich
> Sent: Thursday, June 16, 2016 2:56 PM
> To: dev@openvswitch.org
> Subject: [ovs-dev] [RF
> -Original Message-
> From: Jan Scheurich [mailto:jan.scheur...@ericsson.com]
> Sent: Tuesday, June 21, 2016 10:27 AM
> To: Fischetti, Antonio ;
> dev@openvswitch.org
> Subject: RE: [RFC Patch] dpif-netdev: Sorted subtable vectors per
> in_port in dpcls
>
> &g
Thanks Daniele,
in this patch we moved the reset of batches[i].flow->batch
from dp_netdev_input__() into packet_batch_per_flow_execute().
So before calling dp_netdev_execute_actions() the flow->batch is already NULL.
I think this should prevent recirculation, or am I missing some detail?
Antonio
Thanks Jarno, will follow your suggestions. That's really a good improvement
and will help a lot to explain the function details.
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Jarno
> Rajahalme
> Sent: Friday, October 7, 2016 10:08 PM
> To: Bo
Comments inline.
Thanks,
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Jarno
> Rajahalme
> Sent: Friday, October 7, 2016 10:08 PM
> To: Bodireddy, Bhanuprakash
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH 01/12] dpcls: Use 32 pack
tches[k]
+of the 1st call to dp_netdev_input__(). */
for (i = 0; i < n_batches; i++) {
batches[i].flow->batch = NULL;
}
From: Daniele Di Proietto [mailto:diproiet...@ovn.org]
Sent: Monday, October 10, 2016 1:31 AM
To: Fischetti, Antonio
Cc: Jarno Rajahalme ; dev@openvsw
Thanks for your comments Daniele, my replies inline.
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele Di
> Proietto
> Sent: Friday, October 14, 2016 3:08 AM
> To: Bodireddy, Bhanuprakash
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev]
Thanks Jarno, one question below.
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Jarno
> Rajahalme
> Sent: Friday, October 14, 2016 5:03 PM
> To: Bodireddy, Bhanuprakash
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH v3 03/12] flow:
Thanks Jarno for spotting this, will do the suggested change.
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Jarno
> Rajahalme
> Sent: Tuesday, October 18, 2016 6:26 PM
> To: Bodireddy, Bhanuprakash
> Cc: dev@openvswitch.org
> Subject: Re: [ovs
Hi All,
Here's an optimization idea for the datapath classifier table.
I'd like to get some feedback.
I used the DPDK ACL tables. They can perform a wildcarded matching and each
lookup requires less CPU cycles than the Classifier.
Anyway there's a negative aspect with ACLs. They take a very long
Hi Zoltan, thanks for your questions.
Please find below my answers inline.
> -Original Message-
> From: Zoltan Kiss [mailto:zoltan.k...@linaro.org]
> Sent: Thursday, December 17, 2015 2:33 PM
> To: Fischetti, Antonio; dev@openvswitch.org
> Subject: Re: [ovs-dev] Wi
> -Original Message-
> From: Zoltan Kiss [mailto:zoltan.k...@linaro.org]
> Sent: Friday, December 18, 2015 12:38 PM
> To: Fischetti, Antonio; dev@openvswitch.org
> Subject: Re: [ovs-dev] Wildcard Matching optimization idea
>
>
>
> On 17/12/15 16:23, Fisch
Hi Chandrakant,
You may want to have a look at the OVS/utilities/ or OVS/vswitchd/ folders.
For ex.
OVS/utilities/ovs-vsctl.c
contains the implementation of commands like 'ovs-vsctl add-br ...'
See 'vsctl_commands[]' structure, it lists the required format, the params check
and the execution of th
Thanks Zoltan for your questions.
Anyone else has any new feedback?
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Fischetti,
> Antonio
> Sent: Friday, December 18, 2015 3:44 PM
> To: Zoltan Kiss; dev@openvswitch.org
> Subject: Re:
ffered
in a 'waiting list' to be processed later.
Background table updated? Go to Step #3.
3. Done
--
Both tables are synced up.
Is there a new rule to insert and/or anything in the 'waiting list'? Go to Step
#1.
> 在 2016/1/12 18:37, Fischetti, Antonio 写道:
> >
Hi Daniele,
I reviewed your patches and they look ok.
I also made some regression tests on performance with an IXIA generator. I
would like
to share these results.
I did the following. I applied the patches and ran a couple of tests without
using the
new 'ct' action, so the ConnTracker feature
functions.
It's strange because with these 3 patches none is calling 'parse_dl_type()'.
If I comment that function, performance goes up again to 12.1Mpps.
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Fischetti,
> Antonio
As this Regression Test issue is specific for Patch #3/12
I'll continue this topic on the post for [Patch 03/12] at
http://openvswitch.org/pipermail/dev/2015-November/062230.html
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Fischetti
Hi, I was running some Regression Test to check performance in a
Phy-2-Phy test, ie
ovs-ofctl add-flow br0 in_port=1,action=output:2
After applying patches #1 and then #2 the performance
is still ok.
Instead, after I apply this patch #3 the throughput goes from 12.1
down to 11.6 Mpps.
I found ou
n emc_processing()
* conn_not_found() 2%
* parse_dl_type() 1%
* lib/conntrack-tcp.c:tcp_valid_new() 1%
Antonio
> -Original Message-
> From: Fischetti, Antonio
> Sent: Monday, February 1, 2016 11:09 AM
> To: Fischetti, Antonio; Daniele Di Proietto; dev@openvswitch.o
)
8% dp_netdev_process_rxq_port()
3% netdev_dpdk_rxq_recv()
...
> -Original Message-
> From: Daniele Di Proietto [mailto:diproiet...@vmware.com]
> Sent: Monday, February 1, 2016 7:38 PM
> To: Fischetti, Antonio
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH 03/12] flow: Introduce par
tl add-flow br0 dl_type=0x0800,nw_tos=2,actions=output:2
sudo ./utilities/ovs-ofctl add-flow br0 vlan_tci=0,actions=output:2
so that the classifier was using 6 Subtables.
Antonio
From: William Tu [mailto:u9012...@gmail.com]
Sent: Wednesday, February 3, 2016 11:52 PM
To: Fischetti, Antonio
Cc: dev@
Hi Jarno,
Below my reply.
> -Original Message-
> From: Jarno Rajahalme [mailto:ja...@ovn.org]
> Sent: Thursday, February 4, 2016 10:00 PM
> To: Fischetti, Antonio
> Cc: u9012...@gmail.com; dev@openvswitch.org
> Subject: Re: [ovs-dev] Wildcard Matching optimization idea
50 matches
Mail list logo
