Re: [lwip-users] Duplicate ACK issue lwIP 1.4.0

[Tim King]

Enrico,

Thanks for spotting that.  I believed my system was thread safe but that
packet indicates it may not be.  I will have a look further and see if I can
find the memory leak.

Tim

-----Original Message-----
From: address@hidden
[mailto:address@hidden On
Behalf Of address@hidden
Sent: Monday, 9 November 2015 10:50 p.m.
To: address@hidden
Subject: lwip-users Digest, Vol 147, Issue 16

Send lwip-users mailing list submissions to
        address@hidden

To subscribe or unsubscribe via the World Wide Web, visit
        https://lists.nongnu.org/mailman/listinfo/lwip-users
or, via email, send a message with subject or body 'help' to
        address@hidden

You can reach the person managing the list at
        address@hidden

When replying, please edit your Subject line so it is more specific than
"Re: Contents of lwip-users digest..."


Today's Topics:

   1. Re: Duplicate ACK issue lwIP 1.4.0
      (Enrico Murador - Research & Development - CET)
   2. Re: Duplicate ACK issue lwIP 1.4.0
      (Enrico Murador - Research & Development - CET)


----------------------------------------------------------------------

Message: 1
Date: Mon, 9 Nov 2015 10:29:49 +0100
From: Enrico Murador - Research & Development - CET
        <address@hidden>
To: Mailing list for lwIP users <address@hidden>
Subject: Re: [lwip-users] Duplicate ACK issue lwIP 1.4.0
Message-ID: <address@hidden>
Content-Type: text/plain; charset="windows-1252"; Format="flowed"

Hi,
I gave only a quick look at the log... I stopped at packet 167, that is a
malformed part of a TCP ECHO response sent by Xilinx board. This message has
some fields not OK: Ethernet frame type, IP version field, and an IP service
field. All of these are located in a 32bit word starting at 0x0C from start
of message, so you have an 0x40000000 instead of 0x08004500 at that
location.

I could say there are some problems with multithreading and/or with your
ethernet low level driver, but you probably already checked these parts. 
Maybe some other function in your system is "simply" writing a longword in
memory at the wrong place?
Don't know if this issue could be related with duplicate ACKs (surely it is
related with the "TCP previous segment not captured" at packet 168), but I
would try to resolve it first.

Enrico

On 06/11/2015 10:16, address@hidden wrote:
> Hi,
> My system is on a Xilinx Microblaze.  I am running lwIP 1.4.0 under 
> FreeRTOS.  I am using it in raw mode with a single tcpip thread.
> I have an issue where I am getting duplicate ACKs and the connection 
> is failing.  I am not sure what exactly is the cause but Wireshark
> (attached) shows duplicate ACKs.  lwIP is on port 7 @ 192.168.1.10.
> I have tried different settings and get minor changes.  This occurs 
> during a high traffic loading sequence.
> I have set the lwIP task at the highest priority and it loads 95% of 
> the time but still occasionally gets the issue.  However I don't wish 
> it to be the highest priority as there are more important (but with 
> low computational effort) tasks.  When I lower it to a mid priority 
> task it loads maybe 5% of the time.
> I have DDR memory in the system and have a large heap and have 
> allocated a lot of pbuf memory.
> Anyone have any ideas on what might be causing my issue and 
> suggestions on how to fix/improve it?
> Thanks,
> Tim
>
>
> _______________________________________________
> lwip-users mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/lwip-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.nongnu.org/archive/html/lwip-users/attachments/20151109/a02116
69/attachment.html>

------------------------------

Message: 2
Date: Mon, 9 Nov 2015 10:49:39 +0100
From: Enrico Murador - Research & Development - CET
        <address@hidden>
To: Mailing list for lwIP users <address@hidden>
Subject: Re: [lwip-users] Duplicate ACK issue lwIP 1.4.0
Message-ID: <address@hidden>
Content-Type: text/plain; charset="windows-1252"; Format="flowed"

...and if it could help (and if I'm not wrong) that packet was correctly
retransmitted (packet 188), so maybe packet 167 has effectively been altered
after creation.

On 09/11/2015 10:29, Enrico Murador - Research & Development - CET wrote:
> Hi,
> I gave only a quick look at the log... I stopped at packet 167, that 
> is a malformed part of a TCP ECHO response sent by Xilinx board. This 
> message has some fields not OK: Ethernet frame type, IP version field, 
> and an IP service field. All of these are located in a 32bit word 
> starting at 0x0C from start of message, so you have an 0x40000000 
> instead of 0x08004500 at that location.
>
> I could say there are some problems with multithreading and/or with 
> your ethernet low level driver, but you probably already checked these 
> parts. Maybe some other function in your system is "simply" writing a 
> longword in memory at the wrong place?
> Don't know if this issue could be related with duplicate ACKs (surely 
> it is related with the "TCP previous segment not captured" at packet 
> 168), but I would try to resolve it first.
>
> Enrico
>
> On 06/11/2015 10:16, address@hidden wrote:
>> Hi,
>> My system is on a Xilinx Microblaze.  I am running lwIP 1.4.0 under 
>> FreeRTOS.  I am using it in raw mode with a single tcpip thread.
>> I have an issue where I am getting duplicate ACKs and the connection 
>> is failing.  I am not sure what exactly is the cause but Wireshark
>> (attached) shows duplicate ACKs.  lwIP is on port 7 @ 192.168.1.10.
>> I have tried different settings and get minor changes.  This occurs 
>> during a high traffic loading sequence.
>> I have set the lwIP task at the highest priority and it loads 95% of 
>> the time but still occasionally gets the issue.  However I don't wish 
>> it to be the highest priority as there are more important (but with 
>> low computational effort) tasks.  When I lower it to a mid priority 
>> task it loads maybe 5% of the time.
>> I have DDR memory in the system and have a large heap and have 
>> allocated a lot of pbuf memory.
>> Anyone have any ideas on what might be causing my issue and 
>> suggestions on how to fix/improve it?
>> Thanks,
>> Tim
>>
>>
>> _______________________________________________
>> lwip-users mailing list
>> address@hidden
>> https://lists.nongnu.org/mailman/listinfo/lwip-users
>

--
Enrico Murador
R&D

CET Electronics logo <http://www.cet-electronics.com/>  *CET Electronics 
snc*
via E. Fermi, 1
31050 Zenson di Piave (TV) Italy
Phone: +39 0421 344100
Fax: +39 0421 464042
PI/VAT IT01780330260
Web: http://www.cet-electronics.com/




-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.nongnu.org/archive/html/lwip-users/attachments/20151109/14ee92
ca/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: logo.png
Type: image/png
Size: 8747 bytes
Desc: not available
URL:
<http://lists.nongnu.org/archive/html/lwip-users/attachments/20151109/14ee92
ca/attachment.png>

------------------------------

_______________________________________________
lwip-users mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/lwip-users

End of lwip-users Digest, Vol 147, Issue 16
*******************************************