Apparent Bitfile corruption in VI

[Intaris]

I was writing a small Host VI to test certain FPGA problems we've been having in the last time.

Test was working fine, and I was refining things step by step.  Then I started getting errors "If using external clocks blah blah blah" and was getting errors writing controls in the FPGA code but reading the indicators was working fine.  I tried restarting LV, restarting the hardware and nothing was able to remedy the problem.

I finally switched Bitfile (in my "Open FPGA Reference" node) and suddenly the problems went away.  I changed BACK again and things worked fine.

This is a very worrying behaviour.  I have both VI and Bitfile on a shared network path and the source for the host VI is NOT set to seperate compiled code.  During testing, every 8-12 times I re-run the VI after making trivial changes the Bitfile fails to work as expected.

Seeing how we're trying to nail down errors which are occurring approximately 1% of system restarts, this is a VERY worrying sign.  Anybody got any experience with this?  I'm using LV 2012 SP1 on WinXP running in a Virtualbox VM which in turn runs on a Win7 64-bit machine.  My XP machine has the 3GB switch with USERVA set to 2900.

Shane

[Intaris]

Clarification:

1) Saving the VI (Recompiling) seems to have an X% chance of corrupting the bitfile.

2) Loading the bitfile to the FPGA target repeatedly without recompiling (re-running the VI) seems to work fine over large numbers of iterations

Shane

[LarsR]

Hi Intaris,

probably the failure occurs due running the application on a virtual machine. Problem is that the VM provides a virtual CPU that may cause timing problems.

Best Regards,

Lars

[Intaris]

Is there any more informaiton on this topic.  Such side-effects of running in a virtual PC are unknown to me.

I don't see how the timing problem could occur in this way, but I'm eager to learn.

Shane.

[Intaris]

Any more information on this?  Is this a real problem?  If so then I need to know that my VM is causing the problems.......

[LarsR]

Hi Shane,

more information about the issue could be helpful.

What hardware do you use?

Are you able to post your example project?

Screenshots of failure messages could be helpful.

Regards,

Lars

[Intaris]

As soon as I see the problem again I'll send a copy of the corrupted VI so that it can be analysed to see what's wrong.  I'll also send a copy of a non.corrupted VI as comparison.  How about that?

It doesn't happen VERY often, but as soon as it happens again I'll make sure to post.

Shane.