[tcpm] Re: Discussion on whether the TCP four-wave mechanism can be simplified to three-wave mechanism

[Yoshifumi Nishida <[e-mail]>]

Hi Yang,
I think aggregating the 2nd packet (ACK for FIN) and the 3rd packet (FIN)
is theoretically possible if you want to do it.
--
Yoshi


On Sun, Aug 25, 2024 at 1:02 AM [e-mail] <
[e-mail]> wrote:

> I find the picture，When the server has no data to transmit, can the second
> and third waves be combined to reduce one wave? Because it has this delay
> mechanism, when there is data to be sent, it is sent, and there is no delay
> waiting. If something needs to be sent during the waiting period, can I do
> an ACK confirmation to send the merge?
>
>
> ---- Replied Message ----
> From Yoshifumi Nishida<[e-mail]> <[e-mail]>
> Date 08/25/2024 15:34
> To [e-mail]
> Cc tcpm<[e-mail]> <[e-mail]>
> Subject Re: [tcpm] Discussion on whether the TCP four-wave mechanism can
> be simplified to three-wave mechanism
> Hi Yang,
>
> OK. It seems to me that you're suggesting that when one peer sends the
> final FIN, it can terminate the connection without waiting for the FIN ACK.
> However, if this FIN hasn't arrived at the other end, the other end cannot
> know the peer has terminated the connection.
> My concern here is that the other end would be suspended in this kind of
> situation.
> --
> Yoshi
>
> On Sun, Aug 25, 2024 at 12:09 AM [e-mail] <
> [e-mail]> wrote:
>
>> yes
>>
>>
>> ---- Replied Message ----
>> From Yoshifumi Nishida<[e-mail]> <[e-mail]>
>> Date 08/25/2024 15:05
>> To [e-mail]
>> Cc tcpm<[e-mail]> <[e-mail]>
>> Subject Re: [tcpm] Discussion on whether the TCP four-wave mechanism can
>> be simplified to three-wave mechanism
>> Hi Yang,
>> Are you referring to the 4 way close in TCP? and you want to make it 3
>> way?
>> --
>> Yoshi
>>
>> On Fri, Aug 23, 2024 at 5:03 AM [e-mail] <
>> [e-mail]> wrote:
>>
>>> IEFT Hello, I am a Chinese computer major student, I have an idea is to
>>> study the TCP four wave mechanism, if one party still has data to pass,
>>> continue to open the connection. However, if both parties have no more data
>>> to transmit at a given moment, can it be possible to dispense with one
>>> wave, thus turning four waves into three? (Go straight back to the finally
>>> state at the end of the third time.) For example, is it feasible to go
>>> directly to the last step, that is, to change from the original four waves
>>> to three waves? Is there a theoretical basis for this assumption and the
>>> possibility of practical application?
>>>
>>> Thank you, looking forward to reply!
>>>
>>> I wish all the best!
>>> Yang Zhiyuan
>>> August 23, 2024
>>>
>>>
>>>
>>> _______________________________________________
>>> tcpm mailing list -- [e-mail]
>>> To unsubscribe send an email to [e-mail]
>>>
>>