Real-Time Measurement and Control

cancel
Showing results for 
Search instead for 
Did you mean: 

compiling .out file for VxWorks 6.3 (NI RT 8.5), to cRIO-9012 target

I have a cRIO-9012 on which I was successfully calling a .OUT file which was built with the GNU compiler, for VxWorks 6.1 as included in NI RT 8.2.  We recently upgraded our software to 8.5, both the labview development system and the installed software on the cRIO target, with NI RIO 2.3.1.  This broke the .OUT call - no error message was produced, but the application would simply stop when the call library function node was reached.  I obtained the new version of the GNU compiler (for VxWorks 6.3) and recompiled our .OUT library using the same source code file and Makefile which was used originally to compile the VxWorks 6.1 .OUT file with the new compiler.  The file was compiled without any apparent errors, and I deleted the original .OUT on the cRIO target and ftp'd the new .OUT to the target (/ni-rt/system).  I expected this to be a simple library replacement, but the new .OUT fails in exactly the same manner - the application terminates without error upon executing the call library function node.

Suggestions?


0 Kudos
Message 1 of 4
(4,835 Views)
This is probably not going to be of much help, but have you rebooted the controller after ftp'ing the new .out file?  Are you sure there are not multiple instances of the .out file on the controller? Have you tried creating a helloworld.out with VxWorks 6.3and running it from the shell or LabVIEW?

That's all I can think of for now.

J. Mota.
0 Kudos
Message 2 of 4
(4,827 Views)
Would you be able to attach your vis to reproduce behavior?
 
Regards,
 
Juan Galindo
Applications Engineering
Juan Galindo
Applications Engineer
National Instruments
0 Kudos
Message 3 of 4
(4,573 Views)
Both the host vi and the DLL/OUT contain proprietary IP, so I am hesitant to provide these; however, as an interim measure I am reverting to Labview Real-Time 8.2 to get the system running with the original .OUT file. That at least will give me some time to try and sort out the problem on the new software in the lab. If all else fails, I will post appropriately sanitized code for assistance.
0 Kudos
Message 4 of 4
(4,546 Views)