Rolf Kalbermatter

Where would the paths be located? There exists something like files system redirection in Windows VISTA and higher which will redirect certain paths like C:\Windows\System32 or C:\Program Files and C:\Program Files (x86) to whatever the kernel considers the appropriate location for the current application based on its bitness. So eventhough you ask for C:\Windows\System32 in a 32 bit process, it will be directed to C:\Windows\SysWOW64 when run on a 64 bit OS! It's possible that LabVIEW attempts to be smart when trying to reference a DLL that is defined in the Call Library Node but won't second guess Windows decision when you define the dynamic path.

Well, lots of questions and some assumptions. I created the cdf files for the OpenG ZIP library by hand from looking at other cdf files. Basically if you want to do something similar you could take the files from the OpenG ZIP library, change the GUID to some self generated GUID in there. This is the identifier for your package and needs to be unique, so you can not use that of another package or you mess up your software installation for your toolkit. Also change the package name in each of the files and the actual name of your .so file. When you interactively deploy a VI to the target that references your shared library through a Call Library Node and the shared library is not present or properly installed then you will get an according message in the deployment error message with the name of the missing shared library and/or symbol. If you have some component that does reference a shared library through dlopen()/dlsym yourself then LabVIEW can not know that this won't work as the dlopen() call will fail at runtime and not at deployment time and therefore you will only get informed if you implement your own error handling around dlopen(). But generally why use dlopen() since the Call Library Node basically uses dlopen()/dlsym() itself to load the shared library. Basically if you reference other shared libraries explicitedly by using dlopen()/dlsym() in a shared library you will have to implement your own error handling around that. If you implicitedly let the shared library reference symbols that should be provided by other shared libraries then the loading of your shared library will fail when those references can't be resolved. The error message in the deplyoment dialog will tell you that the shared library that was referenced by the Call Library Node failed to load, but not that it failed because some secondary dependency couldn't be resolved. This is not really different with Windows where you can either reference other DLLs by linking your shared library with an import library or do the referencing yourself by explicitedly calling LoadLibrary()/GetProcAdress(). The only difference between Windows and elf is in the fact that on Windows you can not create a shared library that has unresolved symbols. If you want the shared library to implicitedly link to another shared library you have to link your shared library with an import library that resolves all symbols. On elf the linker simply assumes that any missing symbols will be resolved at load time somehow. That's why on Windows you need to link with labviewv.lib if you reference LabVIEW manager functions but with labviewv.lib being actually a specially crafted import library as it uses delay load rather than normal load. That means a symbol will only be resolved to the actual LabVIEW runtime library function when first used, not when your shared library is loaded, but delay load import libraries are a pretty special thing under Windows and there are no simple point and click tools in Visual Studio to create them. Please note that while I have repeatedly said here that elf shared libraries are similar to Windows DLLs in these aspects, there are indeed quite some semantic differences, so please don't go around quoting me as having said they are the same. Versioning of elf shared libraries is theoretically a useful feature but in practice not trivial since there are many library developers who have their own ideas about versioning of their shared libraries. Also it is not an inherent feature of elf shared libraries but rather based on naming conventions of the resulting shared library which then is resolved through extra symlinks that create file references for the so name only and a so name with major version number. Theory is that the shared library itself uses a so.major.minor version suffix and applications link generally to the .so.major symlink name. And anytime there is a binary incompatible interface change, the major version should be incremented. But while this is a nice theory quite a few developers only follow that idea somewhat or not at all. In addition I did have trouble to get the shared library recognized by ldconf on the NI Linux RT targets if I didn't create the .so name without any version information. Not sure why on normal Linux systems that doesn't seem to be an issue, but that could also be a difference caused by different kernel versions. I tend to use an older Ubuntu version for Desktop Linux development which also has an older kernel than what NI Linux RT is using.

There is no direct way to install binary modules for NI RT targets from the package manager interface. Basically those binary modules need to be currently installed through the Add Software option in MAX for the respective target. One way I found that does work and which I have used for the OpenG ZIP Toolkit is to install the actual .cdf files and binary modules in the "Program Files (x86)\National Instruments\RT Images" directory. Unfortuantely this directory is protected and only accessible with elevated rights which the package manager does not have by default. Instead of choosing to have to start the VIPM with adminstrative rights to allow the copying of the files to that directory I created a setup program using InnoSetup that requests the administrative access rights on launch from the user. This setup program is then included in the VI package and launched during package installation through a post install VI hook. You can have a look at the Open G ZIP Toolkit sources on the OpenG Toolkit page on sourceforge to see how this all could be done. It's not trivial and not easy, but it is a workable option.

.dylib is basically the actual shared library file that contains the compiled object code. It is similar to the .so file on Linux but in a different object format. .framework is a package very much like the .app for an application. It is a directory structure containing several symlink files pointing over a few redirections to the actual binary object file that is the shared library. In addition it can contain string and other resource files for localization and version information. The low level shared module loader works with the .dylib file or the binary object file inside the .framework, but the MacOSX shared library support works on the .framework level, alhough it does crrently still support to load .dylibs directly too. But Apple tries to move everyone to the package format and removes more and more references in the documentation about the low level access and there is a good chance that support for that is eventually depreciated. This is all in an attempt to remove unportable interfaces from the application level in order to make an application work more an more likely on any iOS compatible device.

Double post here!

You use hardware based timing for your tasks. The only way to get this working as you describe is to buy two seperate DAQ boards and run each of the two AI tasks on one of them. There is one timer circuitry on each board for analog input timing and you can't have two tasks trying to use that circuitry at the same time. There is no software trick to make this work as you want. You could modify your requirements and start the two AI channels together and do the trigger detection afterwards in the read data though. You could even configure the AI task to start together with the AO task by making it a slave of the AO clock.

I use LabVIEW for Mac on an iMac regularly. What do you expect to not work? You don't have many hardware IO interfaces on the Mac but running LabVIEW for Windows in a virtual machine won't give you more options for sure. And BootCamp or whatever that is called nowdays likely will be also not a full solution since Windows doesn't come with drivers for every hardware component in a MacBook Pro.

Well I guess two modulo-remainder operations where a big deal back in the early eighties . Apple smartly sidestepped that issue by choosing 1904 as their epoch for MacOS, and yes I'm sure that was not only to be different than Lotus. As to if that was a deliberate decision back then or more negligence by the gals and guys at Lotus will probably never be found out for sure. It may also just have been something they "inheritet" from VisiCalc.

While it's a breaking change to modify this now, the original of writing seconds since january 1. 1904 as a timestamp to SQLLite is truely broken too. So I would investigate if you can demote the current method as depreciated and remove it from the palettes and add a new one that does the right thing. Existing applications using that function will still work as they used too, and still use a broken timestamp while new developments would use the right method. Also document that difference somewhere for anyone wanting to read databases written with the old method to use the depreciated method for reading, but a string reminder to not use it for new development.

LabVIEW's timestamp format is explicitedly defined as number of seconds since midnight January 1, 1904 GMT. There is no reason LabVIEW needs to adhere to any specific epoch. On Windows the system time epoch is number of 100 ns, since January 1, 1601, and Unix uses indeed January 1, 1970, while the MacOS used January 1, 1904 (yes LabVIEW was originally a MacOS only application! ). And as curiosa, Excel has a default time definition of number of days since January 1, 1900, although due to a mishap when defining that epoch and forgetting about that 1900 wasn't a leap year, the real epoch starts on midnight December 31, 1899. But there is a configuration option in Excel to shift this to the MacOS time epoch! It's definitely a good thing that they stick to the same time epoch on all LabVIEW platforms, and since the MacOS one was the first to be used by LabVIEW, it's a good choice to stick with that. If your API has a different epoch you have to provide functions to convert between the LabVIEW epoch and your API epoch and vice versa.

It's generally correct what JKSH writes when the handle containing the handles is allocated by yourself. However if that array of CodecInfo structs comes from LabVIEW there are strict rules that can be observed and must be followed when returning such data to LabVIEW. Anything beyond the number of elements indicated in the array is uninitialized although the actual array handle space may be bigger but never smaller than needed for the number of valid elements. The only exception to this are ampty array handles which can be both either a valid handle with a number of elements equal to 0 OR a null handle itself. So when receiving handles from LabVIEW you should always check for null and depending on that do DSNewHandle/DSNewHClr or DSSetHandleSize. When resizing a handle existing elements need to be resized and appended elements always created. Removed elements when making an array smaller must always be recursively deallocated as they don't exist for LabVIEW anymore once the array length has been readjusted to a smaller size and would therefore create a memory leaks. If you allocate both the array handle as well as the contained handles inside the array in the same code section without returning in between to the LabVIEW diagram it is entirely up to you if you want to use DSNewHandle or DSNewHClr, as the values inside the newly allocated array need to be initialized anyhow explicitly. The latter requires more execution time but may be faster if you need to initialize most elements in there to 0 or an empty handle anyways. Also it may be a little safer when someone later makes modifications to the code as null pointer dereferencing has a higher chance of crashing than accessing uninitialzed pointers, so debugging is easier.

I thought it would be this but that has many shortcomings. The only way to destroy an occurrence is by calling the DestroyOccur() C API in LabVIEW. However if you do that with an occurrence that was created with the Create Occurrence node that occurrence is gone and the only way to get it back is by unloading and reloading the VI that contains the Create Occurence node. Without reloading the VI, this occurrence will be invalid and immediately run into the timeout case again if you restart the VI. Not exactly a convinient thing when you are developing and starting and stopping your app repeatedly. Of course if you create the occurrence by calling the AllocOccur() API this is not an issue but then you call already two undocumented C APIs.

You forgot a smiley there after the last sentence!

The Create Occurrence node only executes at VI load time and that is by design. It has been that way since the inception of occurrences in LabVIEW back around LabVIEW 2.0. There is the undocumented LabVIEW manager function AllocOccur() that returns a unique occurrence refnum every time it is executed. However since around LabvIEW 6 you have notifiers, queues, and semaphores which internally do use occurrences for the signaling aspect but have extra advantages of allowing to transport some data along with the signal event that occurrences don't have. Occurrences did have a nasty feature in the past and may still have that could bite you if you are not careful. Their state would remain triggered once they were set if the Wait on Occurrence wasn't executed before the program was terminated and would immediately trigger the Wait on Occurrence on a new start of the application even if in this new run the occurrence hadn't yet been set. As to destroying the occurrence to let the Wait on Occurrence terminate, that sounds like a pretty brute force approach. Why resolve to that instead of just setting the occurrence anyways?

You launch the VI inside your RT app and want its front panel to show on your host computer? That is not possible to the best of my knowledge. Your host application can have a front panel that it shows and remotely call your VI through VI server to execute some code on your RT taget, though.

I'm not sure where you see hostility in those remarks. Yes it may be not sugar coated sweet talk, but hostile? I think you should reconsider this.

I take an issue with that n**i word. If someone comes to you and tells you your application doesn't run, you would also try to educate him that there is a lot more information needed in order for you to mean something for him, wouldn't you? In this case it is not even software I wrote so how would we have been able to even guess from the first two posts of the OP that GPIB might be involved? Nor did he provide any information about his application other than that it caused a specific error number. Posting in OpenG made me actually guess at first it may be related to some OpenG toolkit function, otherwise I had left the post entirely alone in the first place.

Well, it is already pretty helpful to know that the function uses GPIB, but it would be even more helpful to see what is actually inside that VI. Most likely it uses the old GPIB function interface and there error 6 means actually that the GPIB IO operation was aborted. That is a somewhat generic error for GPIB operations that the GPIB controller detected some kind of error and has aborted the transfer because of that. Check out this document for a list of possible GPIB errors and what it could mean. As to posting a photograph of a screen shot, well, posting the actual VI AND subVI would have taken less work for you and would be about 100 times more useful to be able to see how the VI that causes the error is built up. We still only can guess that it is probably the GPIB function causing that error.

We don't know your ion gauge, nor what it does and how it functions. It could be that the instrument driver for it simply returns its own errors as LabVIEW errors so that the actual numeric value means something completely different than what one would expect from LabVIEW functions. It can also try to read some file somewhere. All guess work if you don't provide a LOT MORE information about your program, hardware and source code. You usually don't go to a car shop and tell the mechanic that your car doesn't work and if he could tell you what the problem is without taking the car with you so he can have an actual look at it, do you? You know more about your software than anyone here does, so you need to tell us as much as possible about it in order to allow us to help you! As to where to post, I notified a moderator and they moved the thread into the General area, where you should have posted it in the first place.

I didn't mean external hardware but what is build into the system. A memory module, a built in HD, the system board. Together with a less than perfect thermal design that can mean that the hardware gets hot enough that some less than perfect hardware gets into soft errors.

Definitely not an Instrument Driver thing. The only thing besides faulty hardware that can still create BSODs in Windows are faulty device drivers. They run in the kernel space of Windows and there is really nothing that Windows can do to prevent a kernel driver from corrupting its memory. That is why 64 Bit Windows (and the latest versions of MacOS X) by default only allow to load signed drivers. In order to get a driver signed the manufacturer has to submit it to some test review process and that tries to make sure the driver conforms to certain test scenarios in order to make sure it will work flawlessly in all but the very most extreme exceptions. Today a BSOD is a pretty reliable sign that you have some hardware problem in your system, such as a bad harddisk, PCI bridge, or memory that can cause temporary dropouts.

Error 6 is a permission error. That can have many reasons and without knowing which function gives you this error there is really not much we can do to help you. Most likely it is a file permission error. So maybe you login with a different user than before or your system administrator changed the access rights for your user account. By the way: You posted in the wrong forum too. This seems to have nothing to do with OpenG functionality at all, so should have been posted in one of the more general forums.

One possibility is to use the Call Chain and reference the latest element in the resulting array. That is the top level VI of the current call chain, so might not be the top level of the entire application (when you use VI Server->Run VI or Call Asynchronous, but should be close enough). Otherwise there should be an "App->Parent Window for Dialogs" property but last time I checked that always returned a 0 handle, which for most Windows APIs requiring a parent handle would mean that it is system modal as the 0 HWND is treated as a shortcut for the Desktop window, which is the parent of any other window on the desktop.

That's pretty harsh! The source code is on sourceforge and there is nobody preventing other people from accessing it and attempt to port it to 64 Bit LabVIEW. It won't be easy but given enough determination it is certainly doable.

It might be possible but it is far from just a recompilation of the code. The code was written at a time where nobody was thinking about 64 bit CPUs and no standards existed how one should prepare for that eventualiity. Also it is possible that the script interface in LabVIEW has been cleansed of support for older API standards for the 64 bit version. LabPython uses the first version of that API, but if the 64 bit version of LabVIEW doesn't support that anymore, then the new version documentation would need to be gotten from NI. This is not an official public API.