If the communication needs to be 2-way then I would suggest that each system (you seem to be describing 2 but there could be more) have its own private output file which is read by the other(s). Each system writes its 'output' info to the file which would be seen as 'input' to the other(s). This avoids problems with 2 or more programs trying to write to the same file. -- Harvey

Dont apologise. I have been on to Maunder for a while to create a kernel forum. :) First, eliminate the warnings. Tfhis is the kernel here, we accept nothing but 100% perfection. :) (And seriously, I mean that. It is either 100% or it is crap. Nothing in betweem.) So, ewhen the inf checks out OK and rins you need to check the services key to make sure the driver is there. If not, the inf failed in some way. Look at setupapi.log file, it will tell you about inf based installs. If the services entry is there, we need to look at signing, turn off driver signing checking, else the system wont load it. Slight aside here, drivier signing checking on 64 Vista+. Originally you couldnt load unsigned drivers, then the driver dev world got pissed of with MS, and they changed it, but you have to be booted in debug mode (last I heard, but it is possible they changed it on later OSs. Check on google). It coule be your driver is loading, but failing driver entry. So, after it loads (supposedly), do a sc query and see if it says running. If not, it is time to hook up windbg and set a breakpiont on your driver driverentry. bp !driverentry Load it, and it should break. (set the symbols up.) Then you can see if it is failing in driverentry. If it suceeds driverentry, your driver IS running. If your service entry is there, but driverentry isnt getting hit, then you need to make sure it is compiled properly, it needs a check sum, so make sure you use the ddk build utility. Yep, its PITA, but thats why I get paid so damn much! :) ============================== Nothing to say.

Thanks Topcoder for the links

Consider: what is the difference between user space and kernel space? Also consider whether this really has anything at all to do with Managed C++/CLI. Unrequited desire is character building. OriginalGriff

You'll have to add Imports System.Runtime.InteropServices Imports Microsoft.Office.Interop.Excel

Aescleal wrote: Pointer and array questions are on my list of danger sign questions Well, they'd both be high on my list of questions to ask. Can you grok pointers? You don't have to be a complete C head and do them in your sleep, but if you can't handle the abstract thought needed, then there's little point in wasting any more of each others time. It's one of those "you either get it, or you don't things". In STL, you still have: iter++; iter->DoStuff (); which is at the least, pointer-esque. Asking about linked lists was a self contained scenario - and again demonstrates a bit of abstract thinking. As for singleton stuff? My last job wrote software dealing with real physical machines. I used a few singletons. I used it rarely, but it was essential to have the concept. I wouldn't penalise someone if they knew the idea, but not the magic word. Thread safety, exceptions, etc would also make it on my list - I hope that reassures you! Iain. I am one of "those foreigners coming over here and stealing our jobs". Yay me!

Thanks Ash for the reply. I tried malloc with INT_MAX and it indeed fails and with your way we can find out how much exactly the memory can be allocated.

thanks chris

thanks niklas

ok thanks cedric

Process/Thread synchronization has to be handled by the programmer. What I mean is that, for any shared resource intended to be protected by a mutex or semaphore, the wait functions[^] must be used on the handles of these objects so that these resources remain synchronized. The wait functions will block access till a mutex or semaphore is release using ReleaseMutex or ReleaseSemaphore respectively. «_Superman_» I love work. It gives me something to do between weekends. Microsoft MVP (Visual C++) Polymorphism in C

See reply under QA - don't post to multiple forums, it just duplicates work and annoys people. Did you know: That by counting the rings on a tree trunk, you can tell how many other trees it has slept with.

Just for my knowledge:).

Thanks Covean.

ok thanks

Great reply. That clears it all. Also find a good description of query at: http://www.ibm.com/developerworks/linux/library/l-system-calls/ This reply is very much similar to the explaination for windows of read/write() as explained in the chapter IO systems of book Microsoft Windows Internals by Mark E. Russinovich and David A. Solomon

yes thanks for the hints.

peterchen wrote: You prefer the term "raw"? I don't need any qualifiers but "managed" seems to be how Microsoft touts all the .NET BS. peterchen wrote: You, OTOH must be a C++ programmer who thinks harder is better. I'm a C++ programmer who doesn't think C# brings anything worthwhile to the party. Just another of Microsoft's little proprietary ploys. Oh, look what Sun did with Java, I bet we can do that, too. ;P You measure democracy by the freedom it gives its dissidents, not the freedom it gives its assimilated conformists.

yes I have changed the file name to .cpp