Yes I have never had a scope problem with RIP before.
That was my suspicion as well thinking it was my build envirionment and
just to confirm I did rebuild both the v2.5_branch and v2.6.0-pre with RT
with a fresh clone. No problems at all!
Looks like I'm going back to the drawing boards.

As far as I can tell, I am indeed in the v2.5_branch. are you seeing
something else?

Concerning the  "../lib/liblinuxcnchal.so.0: undefined reference to
`_rt_shm_alloc", this error occurs when building with the compilation
option OPT := -O0 (for debug). Seems like the good way to start of course.
The error disappears with  OPT := -Os (for release). Which was what I used
to produce the 'after make clean' else the make stops much earlier.
I think the source for this error is in the control.c code. I'll go back.

I have excluded some RT modules that I initially added but were giving
problems. Tried rebuilding again and it is the 3rd error from my initial
post that is persistent for classicladder/module_hal.c.

Find the make output with the corresponding error at the end here :
http://pastebin.com/Yq4ipZuF
- make is in " /usr/src/linux-headers-2.6.32-122-rtai " directory and
looking for these headers which are not there by default  (#include
"unistd.h" ,  #include "stdio.h", #include "stdlib.h").

Ill go back and check and recheck some things.
Thank you for your help Sebastian.


2013/5/7 Sebastian Kuzminsky <[email protected]>

> On 05/07/2013 07:58 AM, Kel H wrote:
>
>> Could there be a problem with the config default 'run in place'? There is
>> already the version 2.5.1 release of linuxCNC installed on the computer so
>> 'run in place' is logical.
>>
>
> RIP (run in place) seems like the right way to go if you want to have
> 2.5.1 (or some other deb version) installed while you're experimenting.  It
> should work well to have an installed version and one or more rip builds
> available all at the same time.
>
> I just tested a build on a Lucid machine running RTAI, with 2.5.1
> installed (via the deb).  Both the 2.5 and master branches built and passed
> runtests, just like they should.
>
> So I think the problem you're seeing is either something about your build
> environment or about the local code changes.
>
>
>
>  - Yes I have build rt on this computer both V2.4 and 2.5.x.
>>
>
> Ok, then it sounds like the build problem is due to one of your local
> changes.  If you check out a clean v2.5_branch without any local
> modifications (possibly in a separate git clone), does it build
> successfully for realtime?
>
>
>
>  - yes all build dependencies are installed (at least no issues after
>> checkbuilddeps command)
>>
>
> Ok good.
>
>
>
>  make 1 (without issuing a 'make clean') : http://pastebin.com/f1sC4NCD
>> make 2 (more complete after make clean) : http://pastebin.com/duYgsGq0
>>
>
> The 'after make clean' build log you pasted is based on the master branch,
> not the 2.5 branch like you said above.
>
> And....  Maybe my french(?) is bad, but i don't see any error message at
> the end of either of the build logs.  Certainly the
> "../lib/liblinuxcnchal.so.0: undefined reference to `_rt_shm_alloc" that
> you reported originally is not there.  Did you paste the whole thing? Did
> it work?
>
>
> --
> Sebastian Kuzminsky
>
------------------------------------------------------------------------------
Learn Graph Databases - Download FREE O'Reilly Book
"Graph Databases" is the definitive new guide to graph databases and 
their applications. This 200-page book is written by three acclaimed 
leaders in the field. The early access version is available now. 
Download your free book today! http://p.sf.net/sfu/neotech_d2d_may
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to