[lwip-users] "flow control" manipulation with lwIP-1.3.2

2011-03-30 Thread Thomas Richter (TCD - DE/Dresden)
Hi, I´m searching for a possibility to manipulate the "flow control" of TCP data handling. I know that there is the parameter "Window" in the TCP header (see http://en.wikipedia.org/wiki/Transmission_Control_Protocol#TCP_segment_structure). My task is to transport TCP data over Ethernet interface

[lwip-users] handling packets with equal sequence number

2011-03-30 Thread Thomas Richter (TCD - DE/Dresden)
Hi, in my tests the following problem occurs: A packet with the same sequence number was sent to lwIP. Wireshark TCP analysis says it was a TCP Retransmission with same sequence number and same data (plus a little bit more data append to the earlier transmitted data). My question: What will be ha

Re: [lwip-users] handling packets with equal sequence number

2011-03-30 Thread Thomas Richter (TCD - DE/Dresden)
Thomas Am 30.03.2011 16:48, schrieb Kieran Mansley: > On 30 Mar 2011, at 15:39, Thomas Richter (TCD - DE/Dresden) wrote: > >> Give lwIP both frames > Yes, just give lwIP all the frames your driver receives off the wire. Your > driver should have no need to parse the frames at all.

Re: [lwip-users] handling packets with equal sequence number

2011-03-30 Thread Thomas Richter (TCD - DE/Dresden)
will be received and of the frame 2 the data "frame 2 - frame 1" (here: "001122...") is given? Sorry, if I have some understanding problems with this topic. Thanks for all. Best regards Thomas Am 30.03.2011 17:08, schrieb Kieran Mansley: > On 30 Mar 2011, at 15:59, T

[lwip-users] using two memory pools for pbuf

2011-04-07 Thread Thomas Richter (TCD - DE/Dresden)
Hi, is there a possibility to handle two memory pools for pbuf´s in lwIP? I am using TSE Ethernet IF with lwIP-1.3.2. In my application I have opened two ports (first for control and second for data handling) at same net IF. My problem is following: if the buffer pool is full then both connectio

Re: [lwip-users] using two memory pools for pbuf

2011-04-07 Thread Thomas Richter (TCD - DE/Dresden)
Hi Kieran, I´m still working at the problematic that I have to inform from one side of the Ethernet connection to another side that RNR (receive not ready = can not handle more data, please wait) occurs. I thought that I can be use TCP buffer handling and if window size is 0 (in TCP header) than t

[lwip-users] add 2 netif for 2 ETH MAC´s to have 2 buffer pools

2011-05-16 Thread Thomas Richter (TCD - DE/Dresden)
Hi, I will add two net interfaces for two Ethernet MAC´s with lwIP version 1.3.2 (or 1.4.0). I hope that this allows me to handle (two) different buffer pools for my different TCP/IP ports. Is it possible with the same lwIP ( TCP/IP stack)? If yes, use the two different netif´s (adding with netif

Re: [lwip-users] add 2 netif for 2 ETH MAC´s to have 2 buffer pools

2011-05-16 Thread Thomas Richter (TCD - DE/Dresden)
t pool is "full" at this "line" ?? - another idea? Best regards Thomas Am 16.05.2011 10:33, schrieb Kieran Mansley: > On Mon, 2011-05-16 at 09:41 +0200, Thomas Richter (TCD - DE/Dresden) > wrote: > >> Is it possible with the same lwIP ( TCP/IP stack)? >&g

Re: [lwip-users] add 2 netif for 2 ETH MAC´s to have 2 buffer pools

2011-05-16 Thread Thomas Richter (TCD - DE/Dresden)
11, schrieb Kieran Mansley: > On Mon, 2011-05-16 at 11:00 +0200, Thomas Richter (TCD - DE/Dresden) > wrote: > >> What do you think, is it a possible way? > It is, but it will double the memory used for packets. It sounds like > the problem you're trying to solve is that co

Re: [lwip-users] add 2 netif for 2 ETH MAC´s to have 2 buffer pools

2011-05-16 Thread Thomas Richter (TCD - DE/Dresden)
river. Hmm... Thomas Am 16.05.2011 12:39, schrieb Thomas Richter (TCD - DE/Dresden): >> ... it might just be best to allocate enough buffers so >> that each connection can have as many as it wants. > Sorry, this doesn't solve the problem. It will shift the problem to a &g

[lwip-users] at ZeroWindowProbe the response is not ZeroWindowProbeAck

2011-05-17 Thread Thomas Richter (TCD - DE/Dresden)
Hi, during some tests with lwIP 1.4.0 I could provoke that the WindowSize is 0. This tells the sender to stop sending. The next sequence by the sender is a TCP_ZeroWindowProbe sequence (some infos to TCP Analyze Sequence Numbers with Wireshark: http://wiki.wireshark.org/TCP_Analyze_Sequence_Number

Re: [lwip-users] at ZeroWindowProbe the response is not ZeroWindowProbeAck

2011-05-17 Thread Thomas Richter (TCD - DE/Dresden)
Hi Kieran, Am 17.05.2011 16:23, schrieb Kieran Mansley: > On Tue, 2011-05-17 at 16:12 +0200, Thomas Richter (TCD - DE/Dresden) > wrote: >> The next sequence by the sender is a TCP_ZeroWindowProbe sequence >> (some >> infos to TCP Analyze Sequence Numbers

[lwip-users] measure transmission errors with lwIP

2012-01-23 Thread Thomas Richter (TCD - DE/Dresden)
Hi, is there any possibility to measure possibly occuring transmission errors? We will do EMC tests with our hardware and will measure if there are any protocoll specific retransmissions, transmission errors (FEC, ARQ, etc.). Thanks for any tips. Regards Thomas -- Thomas Richter .--