Oh, I see! I thought the fabric port and the destination port were the
same. Thanks to correct me! :)


Weiwei


On Fri, Mar 28, 2014 at 2:54 AM, G Jones <[email protected]> wrote:

> The tap start port is just the source port number. You don't need to
> change that. You just connect the dest_port input on the yellow block to a
> 16 bit number of your choice in the simulink design, and the 16 bit number
> doesn't need to be constant.
> On Mar 28, 2014 1:12 AM, "Weiwei Sun" <[email protected]> wrote:
>
>> Hi Glenn,
>>
>> Thanks for your tip! My understanding for your answer is that we can
>> specify design tunable UDP port for the gbe block in simulink.
>>
>> I'm a little bit confused of the tap_start() in the python corr package.
>> It also specified the port number. Could you explain a little bit of which
>> one the eof would read into the gbe yellow block?
>>
>> Thanks!
>>
>> Weiwei
>>
>>
>> On Thu, Mar 27, 2014 at 4:10 PM, G Jones <[email protected]> wrote:
>>
>>> Yes, the UDP port input is latched with the eof signal, so for each
>>> packet you can update the destination port if you want.
>>>  On Mar 27, 2014 6:32 PM, "Weiwei Sun" <[email protected]> wrote:
>>>
>>>> Hi Casperians,
>>>>
>>>> I have a interesting question, and want to seek help. I use the
>>>> ten_GbE_v2 yellow block to send packet to a destination port which I can
>>>> assign in the software. Is there a way to send two streams of data through
>>>> one physic 10 gbe port with different UDP port numbers? In a other word,
>>>> would it possible that two stream of data with different UDP port number
>>>> can share a 10gbe line?
>>>>
>>>> Thanks for your time and attention! I'm not familiar with
>>>> communication. Any suggestions would be greatly appreciated!
>>>>
>>>> Weiwei
>>>> --
>>>> Weiwei Sun
>>>> Radar Remote Sensing Lab
>>>> Department of Electrical Engineering
>>>> University of Washington (Seattle)
>>>>
>>>
>>
>>
>> --
>> Weiwei Sun
>> Radar Remote Sensing Lab
>> Department of Electrical Engineering
>> University of Washington (Seattle)
>>
>

Reply via email to