This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
Re: ethernet performance <TCPIP guru question>
Thank you Andrew. This makes more sense now. I must have missed the
section on lost packets in the Stevens books.
Steve
Andrew Lunn wrote:
>> I did a tcpdump from my Linux machine. I've noticed two things.
>>First, the TCP window size in ECOS isn't sliding during the transfer.
>>This might be confusing the sender, I'm not sure. Also, I see a burst
>>of data packets and ACKS from ECOS which appear good. Then all of a
>>sudden, ECOS starts to ACK to the same packet over and over. After a
>>few more acks, the sequence resumes. Is this normal? Does this explain
>>the slow network performance?
>>
>
> That i think is happening is someone is loosing packets. This has two
> affects:
>
> 1) eCos will keep sending ACKs for the packet before the one that is
> missing.
> 2) The window size will not grow.
>
> Both will result in poor performance.
>
>
>>Here's the dump of the burst sequence running into the ACK sequence.
>>
>>Thanks.
>>
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1507369:1508817(1448) ack 1 win 5792
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1508817:1510265(1448) ack 1 win 5792
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1510265:1511713(1448) ack 1 win 5792
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1511713:1513161(1448) ack 1 win 5792
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: P 1513161:1514609(1448) ack 1 win 5792
>>
>
> I think this packet above is getting lost.
>
>
>>1.1.1.3.ftp-data > 1.1.1.2.55961: . ack 1513161 win 18824
>>
>
> Here is eCos acking up to, but not including the last packet.
>
>
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1514609:1516057(1448) ack 1 win 5792
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1516057:1517505(1448) ack 1 win 5792
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1517505:1518953(1448) ack 1 win 5792
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1518953:1520401(1448) ack 1 win 5792
>>
>
> The client keeps sending since its not reached its window size yet.
>
>
>>1.1.1.3.ftp-data > 1.1.1.2.55961: . ack 1513161 win 18824
>>
>
> eCos has not received the retransmit, so sends the ack again, just in
> case the ack got lost.
>
>
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1520401:1521849(1448) ack 1 win 5792
>>
>
> More data
>
>
>>1.1.1.3.ftp-data > 1.1.1.2.55961: . ack 1513161 win 18824
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1521849:1523297(1448) ack 1 win 5792
>>1.1.1.3.ftp-data > 1.1.1.2.55961: . ack 1513161 win 18824
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: P 1513161:1514609(1448) ack 1 win 5792
>>
>
> Here is the retransmit.
>
>
>>1.1.1.3.ftp-data > 1.1.1.2.55961: . ack 1513161 win 18824
>>
>
> This could happen is the ACK was already in the queue to be sent when
> the retransmit was received.
>
>
>>1.1.1.3.ftp-data > 1.1.1.2.55961: . ack 1513161 win 18824
>>
>
> This is strange. Why send the ack again?
>
>
>>1.1.1.2.55961 > 1.1.1.3.ftp-data: . 1523297:1524745(1448) ack 1 win 5792
>>1.1.1.3.ftp-data > 1.1.1.2.55961: . ack 1513161 win 18824
>>
>
> Here it finally acks the retransmit
>
>
>>1.1.1.3.ftp-data > 1.1.1.2.55961: . ack 1523297 win 8688
>>
>
> and it acks all the data is had queued up until the last data packet.
>
>
> We need the time stamps to get a better idea. Why send that ack again?
> It could be part of a triple ack i suppose. The time stamps will make
> the obvious.
>
> Andrew
>
>
--
Stephen Polkowski
Centaur Technology
Austin, TX
(512) 418-5730
--
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss