g wrong
> assumptions.
> Thanks,
> Daniele
>
>
> Antonio
>
> > -Original Message-
> > From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele Di
> > Proietto
> > Sent: Friday, October 7, 2016 11:46 PM
> > To: Jarno Raja
ajahalme <ja...@ovn.org>; dev@openvswitch.org
Subject: Re: [ovs-dev] [PATCH 05/12] dpif-netdev: Clear flow batches inside
packet_batch_execute.
2016-10-09 0:54 GMT-07:00 Fischetti, Antonio <antonio.fische...@intel.com>:
Thanks Daniele,
in this patch we moved the reset of bat
aking wrong
assumptions.
Thanks,
Daniele
> Antonio
>
> > -Original Message-
> > From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele Di
> > Proietto
> > Sent: Friday, October 7, 2016 11:46 PM
> > To: Jarno Rajahalme <ja...@ovn.org>
>
detail?
Antonio
> -Original Message-
> From: dev [mailto:dev-boun...@openvswitch.org] On Behalf Of Daniele Di
> Proietto
> Sent: Friday, October 7, 2016 11:46 PM
> To: Jarno Rajahalme <ja...@ovn.org>
> Cc: dev@openvswitch.org
> Subject: Re: [ovs-dev] [PATCH 05/12
This patch basically reverts 603f2ce04d00("dpif-netdev: Clear flow batches
before execute.")
As explained in that commit message the problem is that
packet_batch_per_flow_execute() can trigger recirculation. This means that
we will call recursively dp_netdev_input__(). Here's a stack frame:
Daniele had a comment on this, I believe?
> On Oct 7, 2016, at 9:17 AM, Bhanuprakash Bodireddy
> wrote:
>
> There is a slight negative performance impact, by zeroing out the flow
> batch pointers in dp_netdev_input__ ahead of packet_batch_execute(). The
>
There is a slight negative performance impact, by zeroing out the flow
batch pointers in dp_netdev_input__ ahead of packet_batch_execute(). The
issue has been observed with multiple batches test scenario.
This patch fixes the problem by removing the extra for loop and clear
the flow batches