painting of graph cause program crash

[vincent_lee]

LuisG,

        I don't see top axis control in panel editor.  Attached is a snapshot of a graph control for your reference.

Can you point out where to locate the top axis?

Thanks,

Vincent Lee

[LuisG]

Vincent,

You have a good point. In this screenshot there is no top x-axis. I guess this means that you must be using CVI 6.0. I had assumed that you were using 7.0, since the bug you ran into was only introduced in 7.0.

So now, the most likely explanation is that you must have, at some point, upgraded the CVI runtime to version 7.0, either explictly or implicitly (a number of other NI software products also install the CVI runtime engine). And for the purpose of this bug, it is the version of the runtime engine that matters, not the version of the environment.

To confirm this, can you post here the version of your c:\windows\system32\cvirte.dll?

Thanks.

Luis

[vincent_lee]

LuisG,

You may not believe that I am still working on Win98.

Attached is system infromation got from NI MAX on my PC for your reference.

Actually, I can distribute my application with version 6.0 run-time engine, which work well on user's site.

But, if user install or upgrade some kind of utilities like NI-488.2, which will overwrite run-time engine to the new version included in the utilities. The risk  is that user might upgrade to a version which has the problem that we has discussed for weeks without any hints about why it does not work any more.

Thanks

Vincent

[LuisG]

Hi Vincent,

Well... after seeing your latest screen capture I'm a bit stumped. I can only reproduce the crash with the 7.0 runtime engine, which is consistent with what I know of this bug. However, your screen capture indicates that you are running with the 6.0 runtime engine.
I tried it anyway, but I could not reproduce the stack overflow crash with the 6.0 runtime engine. So, there are only two possibilities: the first possibility is that your program is using the 7.0 runtime engine unbeknownst to MAX (can you search your hard drive for any ocurrences of cvirte.dll besides the windows\system folder? If you find none, can you confirm, by right-clicking on the file, that the version of cvirte.dll in windows\system really is 6.0.4.105). The other possibility is that the crash that you're seeing is not being caused by this bug, but by some other factor instead. That would be a pretty big coincidence, since the CVI 7.0 bug I've been talking about is hard to reproduce, and yet your program reproduces it reliably on my machine, and at the same time, I don't see any other crash on my machine. But it's possible nevertheless. If it turns out that you really do not have the 7.0 runtime engine in your system, and you still see the crash, then that must be the only explanation... if that's the case, I'll have to figure out some other way of reproducing your "real" crash in my computer. If that's the case, would you mind sending me a screenshot of your GPF dialog, when you run the test program that you submitted earlier?

Luis

[vincent_lee]

LuisG,

The program work well on my PC. But, it crash on customer's PC who has CVI run-time engine version 7.0.

However, I can fix this bug by working around with the hint that you provide.  I disable the autoscale by calling

SetAxisScalingMode(panel,coontrol,3,VAL_MANUAL,0,99).

Thanks

Vincent Lee