Benjamin Schelte wrote:
However the call to the function netconn_connect always returns
ERR_INPROGRESS, but after that nothing more happens.
Neither delaying the task and processing later, nor directly calling
netconn_write worked (it also returned ERR_INPROGRESS).
Of course it does,
Andrew Foster wrote:
However, I still have an
allocation on reception of data and a free when receiving unsupported
packet types (!IP or !ARP). These would still happen from context of the
low level input and not the tcpip_thread. Should this be restructured as
well?
That should be protected by
Kieran Mansley wrote:
On Tue, 2010-10-05 at 09:48 -0400, Damon, Antoine wrote:
Have you got an idea about this timeout(?).
Sounds like the server is just waiting before closing the socket,
perhaps because it's giving the client a chance to send another request
but doesn't want to leave
Teckinal Greek wrote:
I want to implement two functions *pbuf2buf()* and *buf2pbuf()*. The
reason is that I wanted to implement these two functions for my lwIP
UDP Client code. I have come to know that the lwIP library provides
these additional functions for buffer manipulation.
I don't know
Simon Goldschmidt wrote:
Try adding an include to mem.h to tcpip.c, this is what might be missing. Any
tips and hints on which version you are using would be much appreciated! :-)
This include was added after 1.3.2, so I suggest you try 1.4.0 RC1.
Simon
goldsi...@gmx.de wrote:
This include was added after 1.3.2, so I suggest you try 1.4.0 RC1.
However, there were still some problems compiling with LWIP_TCP==0 or
LWIP_UDP==0, which I just fixed in CVS HEAD. Thanks for the hint.
Simon
___
lwip
Pettinato, Jim wrote:
I don't see anything amiss... but I must admit I am not a FreeRTOS user so
probably not the best set of eyes to be performing code reviews.
Same for me here: there are some tiny things I would have done
differently, but I wouldn't suspect any of them leading to a
Marco Jakobs wrote:
Wireshark is not what makes me headache about the size ... it's the
debug output of LwIP, recorded with the terminal program. This will
grow very fast, and the problem occurs not that often that i can
reproduce it. But i'll set this up ...
Can't you prevent your boards from
Teratux wrote:
I'm having trouble setting up a two threaded application that reads and
writes from an ethernet port. Is there any problem with setting up
something like this?? Is lwip prepared to share resources like the
netconn* struct and is it able to read and write at the same time (
Yoav Nissim wrote:
Hello All,
We believe there might be a problem when calling lwip_ioctl on a UDP socket.
AFAIK, when calling ioctl() with FIONREAD, we should receive the size of
the next unread datagram;
The problem here is that FIONREAD does not appear to be standardized. We
are trying to
Martin Persich wrote:
Hi,
this is Atmel's bug, by my mind.
I would think so, too, as the rest of the initialization code looks fine.
Simon
___
lwip-users mailing list
lwip-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/lwip-users
Adam Fullerton wrote:
Hi,
Has anyone experienced problems with machines running through VMware?
No, I have been communicating with an lwIP devices from WinXP inside a
VM and it works fine.
What I see is lwIP rejecting packets sent from an application within
a virtual machine due to
ncoage wrote:
I am writing TFTP server (by modifying Luminary Micro code) and I wonder how to
implement timeouts in UDP connections (RAW API). In TCP we have a tcp_poll
callbacks. Is there something similar in UDP?
No, but you can use sys_timeout/sys_untimeout instead to create a timer
I saw that one, too, and fixed it 2 or 3 weeks ago. Although I don't
remember if I checked it in to CVS already (I just came back from
holidays so forgive me for forgetting that). The bug was that the httpd
stopped sending after parsing SSI data and only sent again after a
timeout. By fixing
Kieran Mansley wrote:
I think it is better to write a sys_arch update guild.
The release notes for 1.4.0 will contain details of what has changed to
guide updating sys_arch layers. It would make sense to write this as we
go along, but that doesn't always happen, so at the moment CVS head
Zahir Lalani wrote:
I have managed to get snmp sort of working using the code written by
Christiaan Simons. However, it took me a while to get it talking to
GetIF. The problem was using the native C memory heap manager. If I
did a simple request like time-up, it would work. If I tried
BugTraker wrote:
From contrib directory I can say that an argument 'size' is not used.
You mean the sys_mbox_new() implementation for unix and win32 don't use
this argument? That's not a good thing then...
So what it is for?
First, it should be the minimum number of messages that the
BugTraker wrote:
Hi All,
I'm new to lwIP, read wiki about porting lwIP for RTOS but it mentions
cc.h sys_arch.c sys_arch.h files. Where are they?
Example files for porting lwIP to win32 or unix can be found in the
contrib module in CVS or as a ZIP file in the download area.
Should I
Heyu Zhu wrote:
Hi everyone,
I set up a client using lwip-1.3.2 and both sever and client are
connect by ethernet.
When i call lwip_connect the client do send a correct arp-request
packet and the sever
give a correct arp-reply packet. But after that don't send any packet
for a long time ,
Bernhard 'Gustl' Bauer wrote:
I have done so. It compiles without errors. But no packets are comming out.
LWIP_DEBUG says:
etharp_query: Ethernet destination address unknown, queuing disabled,
packet %p dropped
So I enabled ARP_QUEUEING. And now etharp_timer expires.
So far my LWIP did only
Tyrel Newton wrote:
However, if the processor does the final copy (without a DMA
enginge), than it's a bad thing if the data is not aligned. But you
should be able to include a DMA engine in your FPGA, so...
Xilinx provides a gigabit mac with a built-in DMA (at an additional
cost of course),
Tyrel Newton wrote:
As to the aligned pbuf payload: I think the code currently relies on mem_malloc
returning aligned data (and that should be OK with your current settings), so
you might want to check the return values of your libc malloc.
As the pbuf code is written (I think I'm
mgro...@demmel.com wrote:
There are some structures that are by default not packed but only work
if they are packed.
The following structures are:
struct mem in mem.c
struct pbuf in pbuf.h
struct dhcp in dhcp.h
Is there any reason that they are not packed by default? Because I think
they should
Bernhard 'Gustl' Bauer wrote:
I'm running short of ROM, but have plenty RAM. So why don't I zip the
html code (about 300k) during compilation, place it into ROM and unzip
this code into when booting lwip.
What do you think about this?
Good idea, as long as the unzip code is smaller than the
Kieran Mansley wrote:
transmit is only an issue if there is IP fragmentation. Is that
correct? If so, then is it true that I would not have problems if I
have a known MTU and never send a PDU larger than the MTU? Or in
other words, if I never send a UDP datagram that doesn't fit in the
MTU,
Douglas Atique wrote:
Hi,
I have downloaded the lwip-1.3.2 tarball but it seems not to have any
makefiles or any other build artifacts. Is that correct? I haven't
found any step-by-step information on how to build it either.
I need to embed it in a Windows application that will receive
Oscar F:
Hello everybody!!
I´m using EVK1100 with FreeRTOS and lwip, and i would like to know
this question.
When i called the send function, is the packet sent to the ethernet
line or this is queued and the return say me only that situation?
That depends on the API you are using (I'm
Mike Kleshov wrote:
On 22 April 2010 14:46, Simon Goldschmidtgoldsi...@gmx.de wrote:
There's already a working (raw-API-)httpd in the contrib module in CVS, so
unless you have specific needs, there's no need to write your own. And even if
you do, its source is a good example to write
the
answer is... :-)
Simon
goldsi...@gmx.de wrote:
tes_wzm wrote:
I use lwip+ppp to send data to my tcp server.
When sth occured, I want to close tcp thread. How can I do it? If I kill
this thread through OS, it will leak some memory which use to create
mailbox
and sem. Is't some
Lorenzo Tessiore wrote:
err_t
tcpip_apimsg(struct api_msg *apimsg)
{
struct tcpip_msg msg;
if (mbox != SYS_MBOX_NULL) {
msg.type = TCPIP_MSG_API;
msg.msg.apimsg = apimsg;
sys_mbox_post(mbox, msg);
sys_arch_sem_wait(apimsg-msg.conn-op_completed, 0);
The above line
Luca Ceresoli wrote:
goldsi...@gmx.de ha scritto:
Patches belong into the bugtracker to make sure they don't get
forgotten: http://savannah.nongnu.org/patch/?group=lwip
Done. Sorry for the noise, I considered my patch trivial enough
(http://lwip.wikia.com/wiki/Contributing_to_lwIP
Luca Ceresoli wrote:
The attached patch adds const in front of parameters that deserve it, in
ip_addr.*.
Patches belong into the bugtracker to make sure they don't get
forgotten: http://savannah.nongnu.org/patch/?group=lwip
Thanks,
Simon
___
Marco Jakobs wrote:
OK ... good to know. :-) But i see no advantage to rewrite it from the
netconn API to the socket API, as everything works fine
No, just wantet to make this clear for anyone having to decide between
the 2 APIs.
So the calculation of TCP_SND_QUEULEN will result in 8. Does
Marianovich, Andre wrote:
Ok, now I took a design from xilinx with all the necessary interrupt
stuff, etc. and I think it should work now, but it doesn't. It's all the
same like before. The tcp_connect function doesn't seem to call the
callback function, and the transfer_data function will be
Dany Thiffeault wrote:
wow, great thanks. I'll make some tests tomorrow.
Indeed, I forgot to mention. I'm using TCP with the netconn API. While
reading through the multiple posts I missed the past few months, I
found out that 1.3.2 is out and Atmel framework 1.7 too. I'll upgrade
everything
Marco Jakobs wrote:
Am 20.03.2010 22:00, schrieb Kieran Mansley:
Could you not just use a non-blocking write call, and try again later
if it would block as this would mean that the buffers were probably
all in use.
You can either set the netconn to be non-blocking by default, or add
Based on the event-callbacks, you should be able to know when writing
will succeed or not. This has been done already for the socket API:
writing to a non-blocking sockets won't be executed unless there's
enough buffer space available. And you can use select to wait for more
buffer space to
Mykola Kyrylenko wrote:
Looking through threads
(http://lists.gnu.org/archive/html/lwip-users/2009-12/msg00073.html),
the implication was that a version of PPP with no OS was almost ready to
be released (for 1.3.2).
Wondering how it is progressing? I would prefer not to 'customise' the
LwIP
Sirjee Rooplall wrote:
Hi Support,
LOL! That's a good one ;-)
Simon
___
lwip-users mailing list
lwip-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/lwip-users
already.
Simon
Il 04/03/2010 16.13, goldsi...@gmx.de ha scritto:
It seems you are violating threading requirements: calling the netif_*
and dhcp_* functions from any other thread than the tcpip_thread is not
allowed! And if the code below *would* run in that thread context, it
would block RX
Oscar F wrote:
This file has all packets. you have to open with wireshark. If you see
at the end of the file, there are the problem that i told you.
The malformed packet warning is a warning generated by wireshark when
it detects a protocol but can't parse it correctly. In your case, all
Sägesser Walter wrote:
This code actually works, if the server does not close its client it
got at connect time.
But that's not what I really want. The client task may have nothing to
do for a long time. But when there's something to do, it should open a
connection and send whatever is due and
Martin Velek wrote:
The problem is not a standard(C89) compiler, the problem is the
standard sprintf function. As you know the embedded system is limited
in source and full sprintf uses a lot of stack and ROM code. For this
reasons, there are many simple replacement of sprintf but of course
with
Martin Velek wrote:
I think that the most portable are thouse defined in cc.h. Anyway I am
trying to be consistent with
http://lwip.wikia.com/wiki/Porting_for_an_OS but it is quite hard if
the core is not.
Well, I think the wiki should follow the code, not the other way round.
I'm aware
It seems you are violating threading requirements: calling the netif_*
and dhcp_* functions from any other thread than the tcpip_thread is not
allowed! And if the code below *would* run in that thread context, it
would block RX-packets: the tcpip_thread must process incoming packets
from its
Jeff Barber wrote:
It is possible to do this without a dedicated DNS server or DHCP
server, but the client host and your device would need to implement
zero configuration networking (see
http://en.wikipedia.org/wiki/Zero_configuration_networking as a
starting point).
We have a limited
Bill Yang wrote:
I created a empty lwippools.h and copy/paste the content from your email.
So does it work now or not?
As you said the lwippools.h file has been existed in your server, I tried to
get this file from your web page, but I got the error as below.
The requested URL
I guess these checks should be omitted when using the heap instead of pools.
Simon
Martin Velek wrote:
Hi,
there is a sanity check in init.c
#if (LWIP_UDP (MEMP_NUM_UDP_PCB=0))
#error If you want to use UDP, you have to define
MEMP_NUM_UDP_PCB=1 in your lwipopts.h
#endif
#if (LWIP_TCP
Chen wrote:
What is the difference between netif-input and ethernet_input? Both
seems to work --- in fact, Atmel was using ethernet_input in their
port of lwIP for a long time and the example seems to work OK.
netif-input is the function passed to netif_add. For NO_SYS==1, there
is no
Yigal Hochberg wrote:
I used the the lwip cvs code for the lwip ipstack + winpcap for layer2 ethernet
+ w32 sys_arch.c
of contrib.
I would like to be able to test code changes before I generate a patch.
This way I am sure to submit compiled code as well as unit tested code.
The Windows
Christian Steffen wrote:
Hello,
I use the lwIP/FreeRtos at the Atmel AVR32UC3. The current lwIP Version
I use is 1.3.2.
[..]
Now I see, that the ETHINT task don't use messages to commit the
received frames to the TCP/IP task. It calls ethernet_input() directly.
This is a bug which I think
Bill Yang wrote:
Hi,
After I enabled the MEM_USE_POOLS and MEMP_USE_CUSTOM_POOLS, so I can use the memory
pools. However, when I compiled the lwIP stack used in a demo of freertos, I got a
compiling error said that lwippools.h: No such file or directory as below
message.
Christian Steffen wrote:
Is there any reason, why the global variables not initialized?
Yes: the C standard specifies them to be initialized to zero/NULL and so
it saves space not to do so explicitly.
You have to zero the section of global, uninitialized data in your
startup code to be
Kieran Mansley wrote:
I'm still not quite sure why you would ever need
#if defined(LWIP_SNMP) || defined(LWIP_SNMP_MIB2)
as in your previous example. Surely code in lwIP will either be part of
the SNMP agent protocol (and so controlled by LWIP_SNMP) or be the MIB-2
interface (and so
Mike Kleshov wrote:
I'd like to share my experience with SNMP too.
I had to add SNMP support to my application. I looked at lwip's SNMP
agent and decided that it was too big and complicated for me (I didn't
need all the features.) So I decided to create my own SNMP agent. The
result is a
Yigal Hochberg wrote:
mib-2 includes the system-group as well. any agent will want it.
(btw: rfc1213 is gone).
the scope of discussion are mib-2 objects related to the ipstack.
therefore i suggested the flag LWIP_IPSTACK_MIB2
this scopes the following mibs: if-mib, ip-mib, udp-mib, tcp-mib and
Yigal Hochberg wrote:
I sent the email below to Christiaan Simons christiaan.sim...@axon.tv
mailto:christiaan.sim...@axon.tv.
This email is about an optional SNMP-Agent (v1,v2c, and v3) for Lwip
Unfortunately no response from from Christiaan Simons
I didn't hear (or see an email) from
Joe Dupre wrote:
I need to convert an application from the raw API to Netconn (just as I think
I'm getting proficient with the raw API!)
Can someone point me to current sample code demonstrating a Netconn UDP
connection with callback? Setting up the netconn seems pretty straightforward,
but
Sirjee Rooplall wrote:
Can some explain the difference between memp_malloc and mem_malloc, or
point me to any documentation that will explain this to me.
mem_malloc allocates data from a heap, much like the std-C malloc. Since
a heap can be slow (to find a suitably sized free area) and can
Sirjee Rooplall wrote:
The reason for the MEMP question is that I am trying to track a bug
that causes my system to lockup after a particular PPP packet arrives
and pppInProc processes it, then passes it ip_input after which it
gets posted into the mailbox using sys_mbox_post, after this I
Cristina Marcucci wrote:
1. My thread reads data with lwip_read (I have set a timeout of 36 seconds
as a socket option, but the problem occurs also without the timeout
directive). After working fine for some time, my thread hangs on the
lwip_read, no more data received or a timeout happening.
Lou Cypher wrote:
And the best I can have is an
Optional dual buffer memories, 4K byte ping-pong,
for Tx and Rx
So buffering at most *two* incoming packets.
Have to confess I never made a deep survey on silicon vendors' choices, at MAC
level, in the various devices (i.e. MCUs).
Taking the
Bill Auerbach wrote:
From: Stephane Lesage [mailto:invalid.nore...@gnu.org]
Compiling with #pragma pack(2) works for me.
As did I and discussed it here:
http://lists.nongnu.org/archive/html/lwip-users/2009-09/msg00027.html
Nothing in lwIP that is byte aligned is accessed by casting to
Kieran Mansley wrote:
On Wed, 2010-02-03 at 09:08 -0500, Chen wrote:
Hi, all,
How do I control the numbers of TCP connection to the same port?
If you mean the number of concurrent tcp client connections to one
listening connection, that would be the backlog, and you can set it with
Bill Auerbach wrote:
This means tcp_listen_with_backlog can be used to enforce only one
connection?
Exactly. That's what the backlog parameter in posix listen() does, also.
If this backlog limit is exceeded, what happens?
Incoming connections will be sent a RST. However, this doesn't
You seem to be needing a netbios name server for your device. Have a
look at apps/netbios/netbios.c in the contrib module in CVS (or the
download area), this should be exactly what you need. To set the name,
you have to define NETBIOS_LWIP_NAME to something (either a constant
string or a
Rich Schermerhorn wrote:
unsubscribe
OK, after telling us all what you want, you can go to
http://lists.nongnu.org/mailman/listinfo/lwip-users and let the words
follow action ;-)
Simon
___
lwip-users mailing list
lwip-users@nongnu.org
Simon St-Pierre wrote:
To all newbies like me, this is what I have done:
- Download _http://nongnu.askapache.com/lwip/contrib-1.3.0.zip_
- Add the netbios.c and netbios.h
- In netbios.c define NETBIOS_LWIP_NAME to the name you want to ping.
- In your code, after /prvlwIPInit//();/ and the
Alain Mouette wrote:
when using DHCP I call netifapi_dhcp_start() ant poll netif_is_up()
until the interface is up.
after some long time, if no connection can be made, I want to start DHCP
again (maybe the cable was diconnected and connected to another
network), what should be the correct
smar...@exalogic.it wrote:
looking the code and enabling the ip debug, I see that a icmp packet
seems to be correctly handled:
ip_input: iphdr-dest 0x4e17a8c0 netif-ip_addr 0x4e17a8c0 (0x17a8c0,
0x17a8c0, 0x4e00)
ip_input: packet accepted on interface A1
ip_input:
IP header:
Bernhard 'Gustl' Bauer wrote:
OOSEQ is turned on.
This fits to your description that you might have missed a packet (from
the wireshark log). So unless you don't have any other problems,
treating pbuf chains correctly will do.
I have a few questions that will help me to understand
Which http server are you talking about? The one included with lwIP does
not support CGIs currently!
I guess you are using a server provided by whoever packed lwIP and
provides it for his hardware? Since we don't know the code of that
server, I'm afraid we cannot help you here. (Unless
Bill Auerbach wrote:
How does the lwIP license work in this case? This WEB server is included with
a port of lwIP and includes the original copyrights and acknowledgements of
lwIP from the original code. In this case, can anyone legitimately/legally use
this http WEB server?
Why not?
Bernhard 'Gustl' Bauer wrote:
Hi,
I made an upgrade from 1.3.0 to 1.3.2 and had a lock at the MEMP STATS
before I connected eth. The values for avail did not meet what I
expected from my defines:
RAW_PCB:0/0/0/0 (MEMP_NUM_RAW_PCB == 4)
UDP_PCB:0/4/0/0 (MEMP_NUM_UDP_PCB ==
Sirjee Rooplall wrote:
I found discussions relating to my problem + it existing in LWIP 1.3.0 in
this link.
http://www.mail-archive.com/lwip-users@nongnu.org/msg05554.html
I'd be very cautious comparing possible bugs here. The above mentioned
issue has been wrong sequence numbers while
Madhusudan Bhat wrote:
I was reading the lwip documents and come to know that IP layer wont
handle the sending and receiving the fragments. Does this feature
supported currently (1.3.2) or still not?
Which document was that? IP fragmentation and reassembly has been
supported for quite a while
Bernhard 'Gustl' Bauer wrote:
I checked my http_recv(). I have 3 different exits:
1: pbuf_free(); tcp_abort(); return ERR_ABORT;
2: tcp_receved(); pbuf_free(); tcp_abort(); return ERR_ABORT;
3: tcp_receved(); pbuf_free(); return ERR_OK;
Is there anything wrong with an exit? Do I need
Just one more thing about the send timeout - I've written it in bug
#28605, too: What do you want to do on a timeout? With the suggested
implementation below, you don't know how much data has already been sent
when send() returns with an error... And there's not a simple solution
for that
Bernhard 'Gustl' Bauer wrote:
Hi,
I'm using LWIP 1.3.0 with the modified CS8900 driver coming with contrib.
First of all, I think the one in CVS might be outdated since it is not
maintained any more (which is why it has been moved to the 'old'
subdirectory). Please have a look at the
Fabian Koch:
I am in the position where I would need SOCKET_SO_SNDTIMEO to have
Send() at least return after a while when e.g. I unplugged the cable.
Now, I understand that sockopt is unimplemented in LwIP.
Not sockopt in general, but SO_SNDTIMEO is not implemented, that's right.
Could I
Yoav Nissim wrote:
I wonder though - if these locks were taken care of, reading and writing
to a single socket could be simultaneous but, would lwIP be full duplex
'on the wire' as you've said?
Well, that depends on your MAC, of course: If you have a DMA-enabled MAC
where you can queue
This issue has already been fixed in CVS.
David Empson wrote:
Mykola Kyrylenkomykola.kyryle...@rfi.com.au wrote:
I have transferred my LwIP stack to 1.3.2. All seems to be going well
(with minimal testing), except I am getting this compiler warning:
Building ...\tcp_in.c
Micael (abc) wrote:
Hi Guys,
I have a problem with init of lwip, in conjunction with FreeRTOS.
The thing is that the example startup looks similar to this;
---8---8---8---8---8
tcpip_init(NULL, NULL);
IP4_ADDR(ip_address, 192,168,1,47 );
Bob Brusa wrote:
Hence the problem is not the chksum, but somewhere there is a loss of
data. Now I am trying to find out where the true contents of the pubs gets
lost. Unfortunately I also am lost: The internal working of the lwip-stack
is not easy to grasp. Best thing would be if someone
Mathias Zenger wrote:
Hi,
I need to use SNMP. Can somebody tell me if the SNMP support of lwIP stack
1.3.0 is working? What does it support (traps / get / put, which version)?
Is there any documentation available?
Of course there is: see doc/snmp_agent.txt. Other than that, the agent
is
Mathias Zenger wrote:
Thanks for the reply. Sorry for haven't checked Christiaan's documentation.
However, I think there is a problem when using a RTOS, right? (since the
agent is based on the raw-API...).
There's no problem using the raw API together with an RTOS. You only
have to make
Please attach wireshark captures in wireshark's own file format, not as
text export or any other format to make sure they are readable.
Simon
Hafiz Bashir wrote:
Hi,
We seem to have a problem when using LWIP 1.3.2 to request a large file
from a remote server. Our target is an ST7101 running
Frédéric BERNON wrote:
Hi group, and Happy New Year !!!
Is anyone know where we could get the contrib-1.3.2 package (I don't find it in
the download page). Else, I will get it from cvs...
Contrib hasn't changed much (or not at all?) from 1.3.1, so it's
probably not worth adding it in the
Yoav Nissim wrote:
[..]
2. Thread A reads data from the socket.
3. Thread B sends data on the socket.
Unfortunatly, using a socket (or netconn) from multiple threads at the
same time is currently not supported by lwIP.
[..]
If so, wouldn't we need a set of locks for all
netconn operations that
Rigoberto Bermúdez wrote:
excellent! but 1.3.1 *is* released, you mean 1.3.2?
Herm, yeah, 1.3.2, not 1.3.1. Sorry to confuse you.
Simon
___
lwip-users mailing list
lwip-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/lwip-users
Rigoberto Bermúdez wrote:
hi!
i'm trying to compile lwip 1.3.1 with:
PPP_SUPPORT = 1
PPPOS_SUPPORT = 1
NO_SYS = 1
the compilation reports:
#error If you want to use PPP, you have to define NO_SYS=0 in your lwipopts.h
is there a way to have ppp without an os?
Currently,
If you want someone to actually analyze the wireshark log, you might
want to attach a pcap file instead of pasting the text (which is not
really readable).
:-)
Simon
Domen Puncer wrote:
Hello!
I'm using rawapi, NO_SYS, all lwip functions are called from the same
thread.
My setup is: lwip
Marco Jakobs wrote:
Hi @all,
i've written a SNTP client function for my project, which is polling
the time from a NTP server.
Did you know there's an SNTP client available for lwIPin the contrib
module, available via CVS? No need to write your own.
This works pretty good, the packets are sent
patelbaroda wrote:
So instead of using
netif_add(MACB_if,xIpAddr,xNetMask,xGateway, NULL, ethernetif_init,
tcpip_input ); netif_set_default(MACB_if );
netif_set_up(MACB_if );
I use,
do_netifapi_netif_add();
You can do it that way, but netif_set_up() is then still called from the
wrong
patelbaroda wrote:
I have one thread for web server, that call netconn_write function. The
other thread is from ethernetif.c that handles the MACB input packets. The
TCPIP thread is handling TCPIP process as defined
sys_thread_new(TCPIP_THREAD_NAME, tcpip_thread, NULL,
TCPIP_THREAD_STACKSIZE,
could pull data from windows COM port,
and different devices could be tesed with lwIP+ PPP.
Lee
On Mon, Dec 7, 2009 at 3:36 PM, goldsi...@gmx.de
mailto:goldsi...@gmx.de goldsi...@gmx.de mailto:goldsi...@gmx.de
wrote:
yueyue papa wrote:
thanks, the previous problem is solved
yueyue papa wrote:
ppp.h should move from
src\netif
===
include\netif
It is better.
Things like that will have to wait until after the release of lwIP 1.3.2.
Simon
___
lwip-users mailing list
lwip-users@nongnu.org
We *do* have documentation for such questions! In case you still don't
find it:
On 22/11/2009, goldsi...@gmx.de wrote:
Excerpt from doc/rawapi.txt:
The callback function will be passed a NULL pbuf to indicate that the
remote host has closed the connection.
Does that make it clear enough
Marco Jakobs wrote:
Hi Simon,
i have seen this, but i'm not using the raw api
Sorry, my fault, didn't see that :-)
- which is also not
recommended in a multithread environment ...
Hmm, I wouldn't put it like that. If you want top speed, the raw API is
your friend. But to many users, it's
yueyue papa wrote:
thanks, the previous problem is solved.
Another trivial problem is
PPP is configered as mutual exclude as LAN feature. In fact, this
requirement is not correct. LwIP could support both.
I know that both PPP *and* ethernet may be used at the same time.
However, modifying
801 - 900 of 1087 matches
Mail list logo
