Does the cRIO FPGA continue to work after the CRIO processor crashes -- how can that be demonstrated?

[Holubtsi]

I would like to use the cRIO FGPA to operate with the backplane modules in a degraded mode in the case that the cRIO processor is not available (diasabled, crashed, frozen, locked-up, broken, ...). Is this possible? How can I create a disabled processor situation to demonstrate the approach (yes, I want the system to crash!).

[Christian_L]

The short answer is that, yes, the FPGA will continue to run its program even if the RT processor is hung or crashed, as long as the backplane is still receiving power from the power supply. Of course this assumes that the operation of the FPGA is not dependent on the RT processor. For example, if the RT VI is continuously sending data to the FPGA for processing or output, then the FPGA would not continue to operate normally as it would not be receiving data from the RT VI.

It is difficult to completely crash the RT processor. In fact most crashes of RT or any OS are most commonly some program routine that has gone into an endless loop consuming all of the processing bandwidth and preventing any other routine from running. One way to simulate this in LV RT is to create a time critical priority VI with a While loop running continuously without any Wait or LoopTimer function. This VI will use all of the processor time and will starve any other loop, VI, or process trying to run on the RT processor.

[JMota]

One more thing you can do to demonstrate this, is to use the RIO Device Setup utility to write your bitfile to the backplane's flash (make sure the bitfile is set to run when loaded, and the backplane is configure to auto-load from flash), and then uninstall all software from the controller and/or restart it in safe mode.