RE: [lwip-users] TCP payload is doubled

[David Shmelzer]

Yes, it does sound like bug #27504, however I applied the patch
(downloaded the latest tcp_out.c) and I still get this behaviour.
When the problem occurs there's a delay of ~4sec, so something is timing
out which is causing a retransmission.
I think the retransmission is a concatenation of two packets, but I have
yet to determine this.
It only happens in the raw tcp code. The netconn() code I have works
fine.
If I revert to 1.3.0 it works. 

Dave


-----Original Message-----
From: address@hidden
[mailto:address@hidden On Behalf Of
David Empson
Sent: Tuesday, October 13, 2009 5:53 PM
To: Mailing list for lwIP users
Subject: Re: [lwip-users] TCP payload is doubled

That description sounds suspiciously like a known bug in 1.3.1, where
TCP segments are incorrectly attached to the unsent queue.

https://savannah.nongnu.org/bugs/?27504

There is a patch attached to the bug report, and Simon checked it into
LWIP a couple of days ago. If you can get CVS-head or apply that patch
to your existing code it will hopefully fix the problem.

----- Original Message -----
From: "David Shmelzer" <address@hidden>
To: "Mailing list for lwIP users" <address@hidden>
Sent: Wednesday, October 14, 2009 9:41 AM
Subject: [lwip-users] TCP payload is doubled


Occasionally, the payload in my TCP return packet sent from lwip is more
bytes than I'm expecting.
It looks like it may be concatenating two copies of the payload.
When it happens, it enters this section of tcp_enqueue():

  /* If there is room in the last pbuf on the unsent queue,
  chain the first pbuf on the queue together with that. */
  if (useg != NULL &&

I'm using 1.3.1 with latest tcp_out.c and tcp.h.
I did not have this problem with 1.3.0.
I'm still debugging, but wanted to get this out there in case it's a
problem with the new changes to tcp_out.c.



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