01-22-2009 10:36 PM
Pawel K
01-23-2009 10:48 AM
Hello Pawel,
Would you be able to submit the application to the NI ftp site, so that we could try to reproduce the problem?
Thanks,
Luis
01-24-2009 08:55 PM
Pawel
01-26-2009 11:28 AM
Hi Pawel,
You can find several examples of panels in multithreaded applications in C:\Users\Public\Documents\National Instruments\CVI85\samples\utility\Threading\*.*. Those should work just fine in Vista, but it probably depends on exactly which actions are happening concurrently within each thread. Let me know if you are able to reproduce the problem.
Luis
02-01-2009 10:02 PM
I found temporary solution.
My application was calling directly Win32 API to create threads and was using Win32 events for synchronization. In particular main thread was waiting for an event from there newly created thread, confirming that the UI panel was created and is ready to accept input.
It seems that I have an issue with Win32 events or timing on VISTA. When I put main thread to sleep for the time adequate to start another thread with UI panel, my application became working.
This is temporary solution as I should find a root cause of the problem. In the meantime I can make my application working on VISTA, and I'm quite sure that there is no issue with DisplayPanel to run in a separate thread.
Pawel
02-03-2009 10:55 AM
I have a CVI7.1 application which I loaded onto a Vista laptop. The program uses multithreading, and worked just fine. Until, that is, the Vista machine accidentally got set to Power Saver mode. Then the threading just didn't want to work at all. Very odd symptoms - but basically the inter-thread messages (including the CVI TCP library callbacks!) just stopped working. Set the machine back to High Performance mode, restarted the program and it all sprang to life again. Curious!
JR
02-04-2009 05:45 PM
02-05-2009 03:48 AM
I don't have a later version of CVI. The program was written to run on XP - I was only trying out Vista out of curiosity, so its not an issue for me. Another related possibility might be something to do with the Dual Core CPU the Vista machine had - maybe one of the cores gets shut down in Power Saver mode; if this happens after threads/event signalling resources have been committed to the powered-down core it is not surprising that strange things could happen.
JR