<?xml version="1.0" encoding="UTF-8"?>
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns="http://purl.org/rss/1.0/" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:syn="http://purl.org/rss/1.0/modules/syndication/" xmlns:admin="http://webns.net/mvcb/">
  <channel rdf:about="http://blog.gmane.org/gmane.network.uip.user">
    <title>gmane.network.uip.user</title>
    <link>http://blog.gmane.org/gmane.network.uip.user</link>
    <description/>
    <syn:updatePeriod>hourly</syn:updatePeriod>
    <syn:updateFrequency>1</syn:updateFrequency>
    <syn:updateBase>1901-01-01T00:00+00:00</syn:updateBase>
    <items>
      <rdf:Seq>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/2007"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/2006"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1998"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1997"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1990"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1988"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1987"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1986"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1985"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1984"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1981"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1974"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1972"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1971"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1969"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1966"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1964"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1963"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1961"/>
        <rdf:li rdf:resource="http://comments.gmane.org/gmane.network.uip.user/1960"/>
      </rdf:Seq>
    </items>
    <image rdf:resource="http://gmane.org/img/gmane-25t.png"/>
    <textinput rdf:resource=""/>
  </channel>
  <image rdf:about="http://gmane.org/img/gmane-25t.png">
    <title>Gmane</title>
    <url>http://gmane.org/img/gmane-25t.png</url>
    <link>http://gmane.org</link>
  </image>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/2007">
    <title>Issue with keep-alive packets in UIP</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/2007</link>
    <description>&lt;pre&gt;Hii,

       I am using the LM3S6965 controller..i have created a client using
freeRTOS+UIP(v7.1.1) stack and IAR as IDE and i am successfully doing a
client-server communication over TCP with server using the BSD socket
interface..the issue is that when server sends keep-alive packets to the
client, the uip stack is not able to respond with an [ACK] packet..due to
this the server resets the connection after sending 2 keep-alive
packets..there is absolutely no response(no outgoing packet) from the uip
stack after the reception of keep-alive packet..i have not been able to
figure out how keep-alive packet is handled by the uip stack(uip.c)? And
why the [ACK] is not being sent by uip after receiving a keep-alive packet?

Thank You.
&lt;/pre&gt;</description>
    <dc:creator>Sushant Chatufale</dc:creator>
    <dc:date>2013-04-04T11:47:21</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/2006">
    <title>Problem sending large TCP packet</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/2006</link>
    <description>&lt;pre&gt;Hi,

I'm using contiki for my master thesis project. I'm now dealing with a
client/server TCP dialog, in which i'm trying to send raw generated data from a
node to my PC, using a netcat on linux as client and a remote node running
contiki as server.

The node take a picture from a camera connected thought an UART, and save the
amount of data in the nvm of the node.
I've succesfully transfer the whole data (5000 bytes) sending TCP packets with
payload of 32 or 64 bytes of data, but when i try to send packets bigger than
64, the client receive only the first packet, send an ACK and the TCP mechanism
start with plenty of retransmission from the server on the node.

I'm supposing that the ack arrives on the node after the TCP retransmission
timeout, and the node start to retransmit the first packet.

I've configured the TCP MSS and WINDOW SIZE in contiki-conf.h to 300.

Thanks for your help

Paolo

&lt;/pre&gt;</description>
    <dc:creator>p.allamano&lt; at &gt;yahoo.it</dc:creator>
    <dc:date>2012-08-18T17:10:30</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1998">
    <title>PSOCK_READ(TO/BUF) doesn't return</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1998</link>
    <description>&lt;pre&gt;Having successfully run a sample httpd app, I’m now trying to run hello-world. From what I understand, it’s different in that it uses protosockets, whereas httpd app doesn’t. In my application callback I have:

static int handle_connection(struct hello_world_state *s) {

  PSOCK_BEGIN(&amp;amp;s-&amp;gt;p);

  PSOCK_SEND_STR(&amp;amp;s-&amp;gt;p, "Response”); // this works

  //PSOCK_READTO(&amp;amp;s-&amp;gt;p, '\n'); // this doesn’t work. If I put it before PSOCK_SEND_STR() I never receive “Response”.
  // PSOCK_REATBUF() behaves the same way.

  //strncpy(s-&amp;gt;name, s-&amp;gt;inputbuffer, sizeof(s-&amp;gt;name));
  //PSOCK_SEND_STR(&amp;amp;s-&amp;gt;p, "Hello ");
  //PSOCK_SEND_STR(&amp;amp;s-&amp;gt;p, s-&amp;gt;name);
 
  PSOCK_CLOSE(&amp;amp;s-&amp;gt;p);
 
  PSOCK_END(&amp;amp;s-&amp;gt;p);
}

I’m very new to uIP. What directions should I investigate in? My goal is implementing a custom protocol over TCP, so the fact that I cannot get a simple echo application to work is a problem.

Thanks,
Dennis.&lt;/pre&gt;</description>
    <dc:creator>Dennis Begun</dc:creator>
    <dc:date>2012-05-23T12:21:49</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1997">
    <title>post online instead of sending emails.</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1997</link>
    <description>&lt;pre&gt;Hi uip-users,

could you please stop sending everyone broadcast emails.
you should post your questins and responses online instead of that.

please respect the rules.

m.u.


&lt;/pre&gt;</description>
    <dc:creator>Mustafa Uzunali</dc:creator>
    <dc:date>2012-05-08T06:39:04</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1990">
    <title>(unknown)</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1990</link>
    <description>&lt;pre&gt;
Greetings everyone:Recently I'm using uip to implement a TCP client,I used uip_connect() to initialize a connection,but it wont work.My code is:
uip_ipaddr_t* ipaddr;uip_ipaddr_t temp;struct uip_conn* conn;int j;ipaddr=&amp;amp;temp;
uip_ipaddr(ipaddr,192,168,1,12);conn=uip_connect(ipaddr,HTONS(5000));
while(1){//main loop}
What could be wrong?Thank you!       &lt;/pre&gt;</description>
    <dc:creator>李兴</dc:creator>
    <dc:date>2012-05-07T12:27:06</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1988">
    <title>Port uIP to MSP430.</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1988</link>
    <description>&lt;pre&gt;Hi where.

Anybody can help me get working uIP on MSP430F5438, I just place together
all needed files and after compilling uIP does not work.

I'm really beginner in this theme, so for the first I just need starting
point to debug.

Technical specs:
MCU - MSPF5438
LAN Controller - CS8900-H board

Thanks for any help.
&lt;/pre&gt;</description>
    <dc:creator>Vasily Fomin</dc:creator>
    <dc:date>2012-02-15T08:49:39</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1987">
    <title>uIP connect issue</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1987</link>
    <description>&lt;pre&gt;Dear all,

I hope this mailing list is still active. I am a new uIP user.

I am working on a simple application running on ATmega328. The embedded 
device should made a TCP connection to the server and first send it a 
communication configuration IP packet. Afterwards the server is 
continuously sending the packet and the embedded device should be only 
properly decoding them.

Unfortunately everything stopped right at the beginning when I attempt 
to connect to the server. For debugging the network traffic I am using 
the Wireshark. Even if the connection is not successful I can still ping 
the embedded device after the connect attempt.

For the reference code I took the microwebserver application 
(http://en.klinkerstein.m-faq.de/index.php/MicroWebServer), that 
compiles nicely using the AVRstudio. The webserver is running really well.

The connection I attempt to perform as suggested in the manual and in 
all examples:

     uip_ipaddr(&amp;amp;VCaddr, 192,168,10,20);
     conn = uip_connect(&amp;amp;VCaddr, HTONS(5005));

Tracking the activity with Wireshark I noticed that instead of 5 or 6 
steps in typical connection sequence there are only 3. And those 3 are 
very different from what they are supposed to be. I attaching two images 
showing Wireshark capture for proper and false connection sequence.

The behavior is just the same whether the connection is initiated in the 
main just after the sei() or within the function, called from the 
uip_TcpAppHubCall. In fact I entirely stripped down the webserver 
application because in the final application I don't need a web server.

I wanted to make sure that the server is working fine and I wrote a 
small test Windows command line application. It is showing that the TCP 
server is behaving as described and expected. So there must be something 
wrong within the embedded device or there is some sort of 
incompatibility of what I strongly doubt. For the moment I am a bit lost 
where to start digging to find the source of this problem.

Every suggestion is therefor welcome.


With kind regards,

Primoz.

&lt;/pre&gt;</description>
    <dc:creator>Primoz Lemut</dc:creator>
    <dc:date>2011-11-02T21:34:33</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1986">
    <title>can uip receive multiple segments in a raw of size smaller than MSS ?</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1986</link>
    <description>&lt;pre&gt;Hello,

 

I am writing an ftp client based on uip (and Contiki). I have the following
issue:

 

-          I have set MSS to 536 (in contiki-conf.h).

-          When the exchange of ftp control messages is over (i.e. TCP port
21), and the data connection is opened, the server starts sending the file. 

-          But, instead of putting the first data packet in a TCP segment of
size 536, it splits it in 2 TCP segments of size 268 = 536/2.

-          Its sends both segments in a raw, without waiting for the
acknowledgement of the first TCP segment.

-          But my ftp client (and uip) sends an ACK for the first one, while
the ftp server is waiting for an ACK for the second segment only, generating
a lot of retransmissions (seen on tcpdump). 

-          Furthermore, it's a waste of RAM since I have allocated a 536
bytes uip_buffer for receiving packets of only 268 bytes .

 

So, I have the following question:

-          Is uip capable of receiving a data stream from a server which
sends multiple TCP segment without waiting an ACK for each segment ? Is
there something special to do ?

-          If not, is it possible to configure an ftp server (or a linux IP
stack) so that is always use the maximum MSS for data packets instead
splitting packets in segments smaller than the MSS size ?

 

Thanks,

Thierry

&lt;/pre&gt;</description>
    <dc:creator>Thierry Didi</dc:creator>
    <dc:date>2011-10-05T20:44:32</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1985">
    <title>Copy of uip patched repository?</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1985</link>
    <description>&lt;pre&gt;Hi -

After running into some issues with uIP, I stumbled upon some messages noting
that a git repo had been created (http://gitweb.aeruder.net/?p=uip.git) for
keeping patches together that had been floating around on the list over the
years since the uip-1.0 release.  I was interested in taking advantage of this,
but I'm finding that the gitweb and git server on the appropriate hosts
provided by Andrew aren't currently working.  I have contacted him separately,
but I'm wondering if anyone might have a clone of this repo around that I could
grab?  I can find various derivative codebases that seem like they've
integrated some of these patches, but some of them have changed layouts or made
other modifications so I'd like things as close to the plain distribution as
possible.

Best.

-jsnyder

&lt;/pre&gt;</description>
    <dc:creator>jbsnyder&lt; at &gt;fanplastic.org</dc:creator>
    <dc:date>2011-09-29T18:28:21</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1984">
    <title>UDP custom packets</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1984</link>
    <description>&lt;pre&gt;Hello guys,

I really need your help. I have an application on a board that receive
"echo" UDP packets (one each 5'') from a computer and response back with
replays, and it works perfectly. The "echo" packets has the same meaning
like, ... are you there?
The thing is that once I receive the first "echo" UDP packet, I want to send
data  UDP packets back to the computer, continuously.
And this is my problem ... read on

In the ethernet.c file I have this

    uip_len = tapdev_read(uip_buf);
    if(uip_len &amp;gt; 0)

inside the famous while loop (as you know). So, basically, it checks for new
data and if so, process it, otherwise, check the timers.

Inside my application that manage the response to the "echo" I have this:

void udp_init(void)
{
uip_ipaddr_t addr;
struct uip_udp_conn *c;
uip_ipaddr(&amp;amp;addr, 192,168,0,113);
c = uip_udp_new(&amp;amp;addr, HTONS(15000));
 if(c != NULL) {
uip_udp_bind(c, HTONS(15001));
}
}

what I have to do is put something in the while loop to send the data
packets (based on a flag) continuously,  after receiving receiving the first
"echo" packet. So, I need to customize the UDP packet. I tried to copy the
previously received uip_buf in a temp variable and packets are send out, but
malformed. I put the memcpy, set the uip_len, ecc. but it still does not
work. Hope you can help me.

Thanks in advance,
Regards,
&lt;/pre&gt;</description>
    <dc:creator>Edgardo C.</dc:creator>
    <dc:date>2011-07-15T15:18:58</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1981">
    <title>fsdata.c generator not working</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1981</link>
    <description>&lt;pre&gt;I am currently working on a project based on the Olimex board LPC-E2124.

As per the information on the main page of uIP I downloaded the
fs_generator.zip which is used to generate the file fsdata.c
The executable inside the zip file doesn't seem to work on my PC. It gives
an error that some root directory is not proper or something of that sort.

Is this because of my system not being compatible with the program provided?
If yes, then is there an alternative, solution to this issue?

The system that I use is running windows 7, 64 bit OS.

Thanks in advance,
Bhushan.
&lt;/pre&gt;</description>
    <dc:creator>Bhushan Talele</dc:creator>
    <dc:date>2011-07-08T11:38:41</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1974">
    <title>UDP problems with IP header</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1974</link>
    <description>&lt;pre&gt;Hello again,

I have a problem with my UDP application. When I send UDP packets from my
pc, I get the following error in uIP:
"invalid version or header length" and using a log I can see that it is 0x46
and not 0x45 as expected for IPV4. IPV6 is disabled in my Linux pc and there
are no macros in uIP enabling it. Any idea how to solve it? thx in advance,

Regards,
&lt;/pre&gt;</description>
    <dc:creator>Edgardo C.</dc:creator>
    <dc:date>2011-06-27T13:28:36</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1972">
    <title>UDP connection</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1972</link>
    <description>&lt;pre&gt;Hello,

Could somebody point me an example of sending and receiving simple UDP
packets with uIP?. I want to establish a UDP communication between a
computer and a PCB.
Thanks in advance,
&lt;/pre&gt;</description>
    <dc:creator>Edgardo C.</dc:creator>
    <dc:date>2011-06-24T07:22:02</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1971">
    <title>Retransmission and uip_split - is this working and how is it working ?</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1971</link>
    <description>&lt;pre&gt;Hello,

i have a question regarding uip 1.0 / Retransmission / uip_split:

I am using uip 1.0 and uip_split on an embedded system. From time to time i get a retransmission signaled by uip. 
Analyzing some wireshark traces, i see that a single packet which is sent by my app is correctly cut into 2 pieces by uip_split,
but then sometimes (i think normal behaviour) one of the two or both packets must be transmitted.

How do you do this correctly ? Is it just enough to call uip_send of the previous packet ?
Or must there be some additional implementation / some checks, which part (or both) must be retransmitted ?
When i always send the big application packet a second time, does uip (or uip_split) handle the sequence counters which
must now fit to the first / second smaller part of big packet ?

When i have multiple connections, must a keep the last sent packet of each connection till i receive the ACK on the given connection ?
Is this correct ?

Many thanks for clarifying !

Best regards,

Martin
&lt;/pre&gt;</description>
    <dc:creator>Martin Maurer</dc:creator>
    <dc:date>2011-05-10T12:38:30</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1969">
    <title>port for OMAP-L138</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1969</link>
    <description>&lt;pre&gt;Hi,

Does anyone know if there is a port of uIP for the  OMAP-L138
(ARM926EJ-S) processor?

Thanks,
dpom

&lt;/pre&gt;</description>
    <dc:creator>Dan Pomohaci</dc:creator>
    <dc:date>2011-05-02T07:12:06</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1966">
    <title>Parse form</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1966</link>
    <description>&lt;pre&gt;Thanks all for the replies.

Jerry, I agree with your approach but was hesitant to change the core
functions within httpd.c.

Daniel, I'll check this out for the solutions used.
&lt;/pre&gt;</description>
    <dc:creator>Rick Drolet</dc:creator>
    <dc:date>2011-04-24T15:50:41</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1964">
    <title>Parse form from get/submit</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1964</link>
    <description>&lt;pre&gt;System: Win7/64; NXP LCP1768 Cortex-M3; CodeRed/Red Suite IDE
uIP: v1.0; Webserver sample from NXP

Hello,
I am new to uip but not to embedded systems.

The problem I am facing is to find a method which allows me to parse the
html data in order to return dynamic results into the form.  Perhaps I am
asking and searching with incomplete information. Attached is a sample form
that I had used with WizNet, where a procedure (proc_http) is used to
extract the network settings and LED control .

At what point can I parse the input?  I would think that this would be a new
group of procedures within httpd.c but there doesn't seem to be any
"how-to". I have tried to use an embedded cgi action, but with this I cannot
'see' the html data that was prior to the cgi tag.

I suppose that what I am asking for is a sample or excerpt that 1) performs
the parsing of header to extract the requested parameters, 2) how to
incorporate this such that the integrity of the code base is not impacted.

Any help would be greatly appreciated!
Rick
&lt;/pre&gt;</description>
    <dc:creator>Rick Drolet</dc:creator>
    <dc:date>2011-04-24T00:46:23</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1963">
    <title>Realview compiler error of uip</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1963</link>
    <description>&lt;pre&gt;In Realview compiler tool kit i am getting  errors like below.
 
 
uip.c(924): error:  #513: a value of type "int" cannot be assigned to an entity
of type "struct uip_conn *"
uip.c(938): error:  #42: operand types are incompatible ("struct uip_conn *"
and "int")
 
--------------------------------------code -------------------------------
found_listen:
  /* First we check if there are any connections avaliable. Unused
     connections are kept in the same table as used connections, but
     unused ones have the tcpstate set to CLOSED. Also, connections in
     TIME_WAIT are kept track of and we'll use the oldest one if no
     CLOSED connections are found. Thanks to Eddie C. Dost for a very
     nice algorithm for the TIME_WAIT search. */
  uip_connr = -1;    // ---------------------------&amp;gt;line 924
  for(c = 0; c &amp;lt; UIP_CONNS; ++c) {
    if(uip_conns[c].tcpstateflags == CLOSED) {
      uip_connr = &amp;amp;uip_conns[c];
      break;
    }
    if(uip_conns[c].tcpstateflags == TIME_WAIT) {
      if(uip_connr == 0 ||
  uip_conns[c].timer &amp;gt; uip_connr-&amp;gt;timer) {
 uip_connr = &amp;amp;uip_conns[c];
      }
    }
  }
  if(uip_connr == -1) {  // ------------------------------------&amp;gt;line938
 what is the meaning of that?
Regards
kamil

&lt;/pre&gt;</description>
    <dc:creator>kamil_bahadir&lt; at &gt;hotmail.com</dc:creator>
    <dc:date>2011-04-22T12:52:50</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1961">
    <title>DAVICOM PHY SETUP</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1961</link>
    <description>&lt;pre&gt;I am attempting to get fibre Ethernet (100BASE-FX) working.
I am using the DM9161EP (not the DM9161AEP) PHY.
I have some EMAC startup code 'supplied by Atmel' - this has a 5 second delay
before it writes to the MII Registers (within the PHY chip). The question is,
does anyone know why there is a 5 second delay? The reason I ask is I need to
rewrite all the registers, so should I do this after the 5 second delay??
I also think that the LwIP code also has this delay.

As I need to change the code, I am thinking of cutting out the 5 second delay,
rewriting all the registers, then polling the 'Link Status' bit until it is
set. I will also perform the reset cycle of the PHY chip, because I think under
certain circumstances, the IC can lockup.

__
Having read the datasheet, I have the following information to share - which
you may find useful

Keywords:
Using the 10/100 Mbps Fast Ethernet Physical Layer TX/FX Single Chip
Transceiver. Fiber Fibre Ethernet 100BASE-FX 100FX 1310nm

OK, found out what you need to get this working..
As always, the devil is in the datasheet – and this is what he told me…
Firstly, From Datasheet DM9161-DS-F05 (September 10, 2008)
Section 8, MII Register Description, Address 2 &amp;amp; 3. ‘EP’ and ‘AEP’
chips have different IDs. Does your driver check them against a list of
‘supported’ IDs?
Then, Address 16, FX/FX Select. The default is TX – For fibre, you need to
clear this bit.
Lastly, and this is a pain, many of the pins of the DAVICOM are read as inputs
on power-up. See Section 5, “LI: Latch input when power-up /reset”. The
state of these pins will overwrite the contents of the “MII Register
Description” – So, if you have slightly changed the circuit and put a small
biasing voltage on one of the inputs that may have been open circuit before, it
will change the contents of the corresponding MII Register!!!!!. This happens
on the Atmel board to the Address 0, bit 10: Isolate flag. The SAM7_MAC.c
always overwrites this register…. BUT there are 7 OTHER registers that can
get overwritten in this way!! The SAM7_MAC.c code does not set these to a
‘known’ state.. It relies on the ‘Latch input on power-up’, which
caught me out.
All registers that are affected are listed in Table 5.1 (column ‘I/O’,
those marked with ‘LI’). Read the description of the registers that is
affected on power-up. Make sure you overwrite the contents of these registers
as part of you PHY init routine - to be safe, simply overwrite all writable
values. See Section 8.14, it tells you what these 8 register (groups) are set
to when you power up.
When you are done, you may wish to reset the ‘PHY state machine’  (Address
16, bit 3).
Jon Newcomb
http://www.jnewcomb.com/cv/&lt;/pre&gt;</description>
    <dc:creator>arm&lt; at &gt;jnewcomb.com</dc:creator>
    <dc:date>2011-04-21T08:53:37</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1960">
    <title>uip webserver, stuck in retransmission when accessed via WLAN</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1960</link>
    <description>&lt;pre&gt;Hello,

I was wondering if anyone would recognize this issue.

I'm using uip+FreeRTOS, implementing http webserver, dhcp and ftp on a small
embedded with pages and data on SD card. Normally it works great and I can
access my pages and data via client software on my laptop. 

The pre-condition is though, that my laptop has to be connected to my
wireless-router (D-Link) by wire. If connected via wireless, my webserver is
not able to process pages bigger than uip buffer of 1500 bytes. After
transmitting the first chunk of data, it keeps requesting retransmission.

As soon as I connect my laptop to the router, by wire, there is no issue and
everything is ok.

Does anyone have an idea of what is going on?
What am I missing in this puzzle..

Best regards
Palsson

&lt;/pre&gt;</description>
    <dc:creator>Gestur Palsson</dc:creator>
    <dc:date>2011-04-12T22:01:58</dc:date>
  </item>
  <item rdf:about="http://comments.gmane.org/gmane.network.uip.user/1957">
    <title>dhcp problem in FreeRtos with uip stack</title>
    <link>http://comments.gmane.org/gmane.network.uip.user/1957</link>
    <description>&lt;pre&gt;Dear all,

i currently try to use the uip stack in the FreeRTOS (V6.1.1) 
environment on a EK-LM3S6965 TI cortex-M3 board.

For that i use the CORTEX_LM3Sxxxx_Eclipse example together with a fresh 
compiled gcc (4.5.2).

If i try to get an IP number from a dhcp server, no dhcp_discover packet 
will be send out, with a fixed IP the normal function of the FreeRTOS 
demo example works with out any problems.

To check my compiler chain, i tried also the enet_uip dhcp example from 
the Stellaris sw_ek 6852. There the dhcp works without any problems.


So which steps i had to do, to send out the dhcp_discover ?

Thanks in advance !

Andreas


&lt;/pre&gt;</description>
    <dc:creator>Dr. Andreas Wassatsch</dc:creator>
    <dc:date>2011-03-31T07:38:59</dc:date>
  </item>
  <textinput rdf:about="http://search.gmane.org/?group=$group=gmane.network.uip.user">
    <title>Search Engine</title>
    <description>Search the mailing list at Gmane</description>
    <name>query</name>
    <link>http://search.gmane.org/?group=$group=gmane.network.uip.user</link>
  </textinput>
</rdf:RDF>
