3/2017 10:42 AM, O Mahony, Billy wrote:
> >>>>> Hi Kevin,
> >>>>>
> >>>>> My 2c below..
> >>>>>
> >>>>> /Billy
> >>>>>
> >>>>>> -----Original Mess
lly wrote:
>>>>> Hi Kevin,
>>>>>
>>>>> My 2c below..
>>>>>
>>>>> /Billy
>>>>>
>>>>>> -Original Message-----
>>>>>> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev
On 11/23/2017 02:31 PM, Ilya Maximets wrote:
> Hi, Kevin.
> Thanks for fixing this.
>
> Just a minor comment:
> Wouldn't it be more visual if we'll implement it in below way:
>
> diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
> index 0a62630..c1c3ed8 100644
> --- a/lib/dpif-netdev.c
> +++ b/l
Hi, Kevin.
Thanks for fixing this.
Just a minor comment:
Wouldn't it be more visual if we'll implement it in below way:
diff --git a/lib/dpif-netdev.c b/lib/dpif-netdev.c
index 0a62630..c1c3ed8 100644
--- a/lib/dpif-netdev.c
+++ b/lib/dpif-netdev.c
@@ -3463,10 +3463,17 @@ rxq_cycle_sort(const voi
t; >
> > > > /Billy
> > > >
> > > >> -Original Message-
> > > >> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> > > >> boun...@openvswitch.org] On Behalf Of Kevin Traynor
> > > >> Sent: Wednesd
t;> From: ovs-dev-boun...@openvswitch.org [mailto:ovs-dev-
> > >> boun...@openvswitch.org] On Behalf Of Kevin Traynor
> > >> Sent: Wednesday, November 22, 2017 7:11 PM
> > >> To: d...@openvswitch.org; aserd...@ovn.org
> > >> Subject: [ovs-dev] [RFC
> -Original Message-
> From: Kevin Traynor [mailto:ktray...@redhat.com]
> Sent: Thursday, November 23, 2017 1:05 PM
> To: O Mahony, Billy ; d...@openvswitch.org;
> aserd...@ovn.org
> Subject: Re: [ovs-dev] [RFC PATCH v2] dpif-netdev: Add port/queue
> tiebreaker to rx
nesday, November 22, 2017 7:11 PM
>> To: d...@openvswitch.org; aserd...@ovn.org
>> Subject: [ovs-dev] [RFC PATCH v2] dpif-netdev: Add port/queue tiebreaker to
>> rxq_cycle_sort.
>>
>> rxq_cycle_sort is used to sort the rx queues by their measured number of
>> cycl
dev] [RFC PATCH v2] dpif-netdev: Add port/queue tiebreaker to
> rxq_cycle_sort.
>
> rxq_cycle_sort is used to sort the rx queues by their measured number of
> cycles.
> In the event that they are equal 0 could be returned.
> However, it is observed that returning 0 results in
rxq_cycle_sort is used to sort the rx queues by their measured number
of cycles. In the event that they are equal 0 could be returned.
However, it is observed that returning 0 results in a different sort
order on Windows/Linux. This is ok in practice but it causes a unit
test failure for
"1007: PMD
10 matches
Mail list logo
