> -Original Message-
> From: Darrell Ball [mailto:db...@vmware.com]
> Sent: Tuesday, August 8, 2017 8:12 AM
> To: Ilya Maximets ; Wang, Yipeng1
> ; ovs-dev@openvswitch.org
> Cc: Heetae Ahn ; Kevin Traynor
>
-Original Message-
From: Ilya Maximets
Date: Monday, August 7, 2017 at 4:54 AM
To: "Wang, Yipeng1" , Darrell Ball ,
"ovs-dev@openvswitch.org"
Cc: Heetae Ahn , Kevin
On 04.08.2017 23:39, Wang, Yipeng1 wrote:
>
>
>> -Original Message-
>> From: Darrell Ball [mailto:db...@vmware.com]
>> Sent: Friday, August 4, 2017 11:35 AM
>> To: Ilya Maximets ; ovs-dev@openvswitch.org
>> Cc: Heetae Ahn ; Wang, Yipeng1
-Original Message-
From: Ilya Maximets
Date: Friday, August 4, 2017 at 7:17 AM
To: "ovs-dev@openvswitch.org"
Cc: Heetae Ahn , Darrell Ball ,
Yipeng Wang , Kevin Traynor
Currently, real insertion probability is higher than configured
for the maximum case because of wrong usage of the random value.
i.e. if 'emc-invert-inv-prob' == UINT32_MAX, then 'emc_insert_min'
equals to 1. In this case we're allowing insert if random vailue
is less or equal to 1. So, two of