[etherlab-users] Using EtherCAT EoE and Build Errors

Thomas Bitsky, Jr. tbj at automateddesign.com
Wed Aug 15 17:28:32 CEST 2012


I don't use the EtherLAB suite either; just the master. Well, I'm
attempting to. :-)

So, you just start put the master into Operation from the command line?
I'll give that a try.

>>  Actually ecrt_master_send_ext isn't exported by cdev

I had to build it as a kernel module.

>>  Can you try starting the master only?

I'll give that a try right now.

Thanks!

On Wed, Aug 15, 2012 at 10:26 AM, Matthieu Bec <mbec at gmto.org> wrote:

> note that's using the master stack only.
>
> I don't use the EtherLAB suite. There could be an issue if it requests the
> master and its internal loop doesn't "send_ext". Actually
> ecrt_master_send_ext isn't exported by cdev so that could indeed be a
> problem.
> Can you try starting the master only?
>
> On 08/15/12 07:48, Matthieu Bec wrote:
>
>> I have an EL6614, besides starting the master (stable-1.5), configuring
>> the network interface and bringing it up, nothing else is needed to get
>> EOE connectivity.
>>
>>
>> On 08/14/12 15:15, Thomas Bitsky, Jr. wrote:
>>
>>> Ok, so I made some progress, I think. I was able to build as a kernel
>>> module and install the module. So long as the virtual interface (eoe0s7)
>>> is not up, the program brings the network into Operation and scans at
>>> 100Hz.
>>>
>>> Then, I issue # ifconfig eoe0s7 up
>>>
>>> About 20 seconds later, the entire system hangs and I need to reboot.
>>>
>>> In the kernel log, I am seeing this: BUG: scheduling while atomic:
>>> swapper/0/0x10000100
>>>
>>> My understanding is that occurred when using semaphores when spinlocks
>>> should have been used in the master callbacks. I changed over the
>>> program to spin locks, but the same thing occurred.
>>>
>>> I'm not sure what I could try other than falling back to 1.4.0 and
>>> seeing if I'm just running into a stable-1.5 bug. I'm following the tty
>>> examples in the directory very closely. Beyond changing the hardware
>>> declarations to match the network, it's the same program.
>>>
>>> Does anyone have an EoE example they could share?
>>>
>>> Thanks!
>>> T
>>>
>>>
>>> On Tue, Aug 14, 2012 at 7:29 AM, Matthieu Bec <mbec at gmto.org
>>> <mailto:mbec at gmto.org>> wrote:
>>>
>>>
>>>     Hello Thomas,
>>>
>>>     it might be a simple network config problem: the IP address of your
>>>     virtual interface shouldn't be the same as your field device. Try
>>>     assigning it something different on the same subnet, e.g.
>>>     192.168.127.1/8 <http://192.168.127.1/8> and check your routing
>>>     table is correct
>>>
>>>     Regards
>>>
>>>     On 08/13/12 22:21, Thomas Bitsky, Jr. wrote:
>>>
>>>         Hello! I'm new to using EtherLAB. I've got the master
>>> installed as a
>>>         service, and it's communicating to the network fine. However,
>>>         I'm having
>>>         trouble getting the EoE feature up and running.
>>>
>>>         I have an Ethernet device in the field with the IP Address
>>>         192.168.127.254 that I need to communicate with through a web
>>>         browser. I
>>>         have an EL6601 that the master is able to see fine. When I put
>>> the
>>>         network into Operation, the RUN light goes steady.
>>>
>>>         I'm running on Ubuntu 10.04. I have not installed any real-time
>>>         extensions; I'm just working with EtherLAB right now.
>>>
>>>         If I execute:
>>>         # ethercat eoe
>>>
>>>         I get a listing for eoe0s7 as a virtual interface, and that it
>>>         is down.
>>>
>>>
>>>         So, I created in /etc/network the file ifcg-eoe0s7 with the
>>>         following
>>>         contents:
>>>
>>>         IPADDRESS=192.168.127.254/8 <http://192.168.127.254/8>
>>>         <http://192.168.127.254/8>
>>>         STARTMODE=auto
>>>
>>>         I restarted the computer.
>>>
>>>         To raise the interface, I enter:
>>>
>>>         # ifconfig eoe0s7 up
>>>
>>>         The virtual interface goes up, and the LINK light on the
>>> EL6601 goes
>>>         solid green.
>>>
>>>         However, I'm unable to ping 192.168.127.254, or get the pages it
>>>         serves
>>>         to pop up. So, I did more reading, and I think I need to do some
>>>         function calls in the program. So, I added to my test program:
>>>
>>>
>>>         void send_callback(void *cb_data)
>>>         {
>>>               ec_master_t *m = (ec_master_t *) cb_data;
>>>               sem_wait(&mutex);
>>>               ecrt_master_send_ext(m);
>>>               sem_post(&mutex);
>>>         }
>>>
>>>
>>> /*******************************__******************************
>>> **__******************/
>>>
>>>
>>>         void receive_callback(void *cb_data)
>>>         {
>>>               ec_master_t *m = (ec_master_t *) cb_data;
>>>               sem_wait(&mutex);
>>>               ecrt_master_receive(m);
>>>               sem_post(&mutex);
>>>         }
>>>
>>>         int
>>>         main(int argc, char **argv)
>>>         {
>>>
>>>         ...
>>>
>>>            // setup callbacks for EoE
>>>            ecrt_master_callbacks(master, send_callback,
>>>         receive_callback, master);
>>>
>>>         ...
>>>
>>>         }
>>>
>>>
>>>
>>>         However, it won't build:
>>>
>>>         durability.o: In function `send_callback':
>>>         /home/tbj/srcroot/durability/_**_src/durability.c:279: undefined
>>>         reference
>>>         to `ecrt_master_send_ext'
>>>         durability.o: In function `main':
>>>         /home/tbj/srcroot/durability/_**_src/durability.c:321: undefined
>>>         reference
>>>         to `ecrt_master_callbacks'
>>>
>>>
>>>         This is my makefile:
>>>
>>>         CC = gcc
>>>         ETHERCAT_TOPDIR = /home/tbj/srcroot/ethercat
>>>         CFLAGS = -I$(ETHERCAT_TOPDIR)/include -g -O2
>>>         LDFLAGS = -L$(ETHERCAT_TOPDIR)/lib/.libs -lethercat -lrt
>>>
>>>         OBJECTS = durability.o
>>>
>>>         durability.exe : $(OBJECTS)
>>>         $(CC) $(CFLAGS) $(OBJECTS) $(LDFLAGS) -o durability.exe
>>>
>>>         %.o : %.c
>>>         $(CC) $(CFLAGS) -c $<
>>>
>>>
>>>         Can anyone tell me what I'm missing? I feel like I'm one step
>>>         away from
>>>         having this working, but I can't find anything else in any
>>>         documentation
>>>         that would lead me to the answer.
>>>
>>>         Thanks in advance for any help.
>>>
>>>         T
>>>
>>>
>>>         ______________________________**___________________
>>>         etherlab-users mailing list
>>>         etherlab-users at etherlab.org <mailto:etherlab-users@**
>>> etherlab.org <etherlab-users at etherlab.org>>
>>>         http://lists.etherlab.org/__**mailman/listinfo/etherlab-__**
>>> users <http://lists.etherlab.org/__mailman/listinfo/etherlab-__users>
>>>         <http://lists.etherlab.org/**mailman/listinfo/etherlab-**users<http://lists.etherlab.org/mailman/listinfo/etherlab-users>
>>> >
>>>
>>>
>>>
>>>     --
>>>     Matthieu Bec                GMTO Corp.
>>>     cell: +1 626 354 9367 <tel:%2B1%20626%20354%209367>      P.O. Box
>>> 90933
>>>     phone: +1 626 204 0527 <tel:%2B1%20626%20204%200527>      Pasadena,
>>>     CA 91109-0933
>>>
>>>
>>>
>>>
>>> --
>>> Thomas C. Bitsky Jr.
>>> Lead Developer and Application Engineer
>>> ADC | automateddesign.com <http://automateddesign.com>
>>> P: 630-783-1150 F: 630-783-1159 M: 630-632-6679
>>>
>>>  ______________________________**_________________
>> etherlab-users mailing list
>> etherlab-users at etherlab.org
>> http://lists.etherlab.org/**mailman/listinfo/etherlab-**users<http://lists.etherlab.org/mailman/listinfo/etherlab-users>
>>
>
>
> --
> Matthieu Bec                GMTO Corp.
> cell:  +1 626 354 9367      P.O. Box 90933
> phone: +1 626 204 0527      Pasadena, CA 91109-0933
>
>


-- 
Thomas C. Bitsky Jr.
Lead Developer and Application Engineer
ADC | automateddesign.com
P: 630-783-1150 F: 630-783-1159 M: 630-632-6679
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.etherlab.org/pipermail/etherlab-users/attachments/20120815/e5ad67a1/attachment-0005.htm>


More information about the Etherlab-users mailing list