This error ended up being caused by an index that was out of range for certain conditions. The IB2 was the second element of that array, and there was a case where that could have been out of the size of the array. I ended up fixing it with an if-then statement to make sure the equation was valid for all potential sizes of the array.

For a fairly large model, I get errors in the initialNonLinearSystem function of "43:warning: assignment from incompatible pointer type [Wincompatible-pointer-types]"

I am running the 64 bit windows version, latest nightly build (OpenModelica-v1.11.0-dev-54-g4ec62d0-64bit.exe). I also tested with the Linux version on a VM, and did not get that error. Seems like this is a compiler bug.

Unfortunately I can't share my code for the error to be replicated, but I can tell you this has a significant amount of nonlinearities.

I'm running OpenModelica through the corba interface, and with every version that starts with 1.11, my program can no longer find the .mat results file. In fact, when I do a manual search for the file in windows explorer, I can't find any .mat files that were created in the time frame where I'm trying to run that version. Any idea what changed from 1.10 to 1.11? I don't have the problem when I run OMEdit - just when I'm running OM through Corba.

I was able to run the diagnostics, and it looks like the part that takes the longest is the performance of FrontEnd:
Performance of FrontEnd: time 4.295/4.735, allocations: 0.6534 GB / -175796800, free: 28.79 MB / 474.9 MB

This is almost half of the compilation time for this particular model. It varies by model, but for the ones I have looked at, this is always a large fraction of the compile time. Does this tell you anything about the potential causes?

Thanks!

I don't understand how to run with the +d=execstat in OMEdit. How do I set that debug flag or the +d=nogen?

This is more of a general question. I'm trying to understand what kinds of things in models make translation/compilation phase take a long time. I have models, for instance, that take 1.5 minutes to compile, when the solve only takes 30 seconds or less. Are there known issues of specific types of models that take longer to translate and compile? Unfortunately I can't share the specific models, which is why I ask the question this way.

Thanks - that helps. I was not aware of the -noeventemit option.

Is there a way to limit the output file size? For example, can we choose to output only the data from every 10ms? With the variable step solver, the time step of the data in the .mat file is variable, but there are times when I would only want to see the data on a certain frequency, to limit the size of the file.

32 bit version is able to open fine.

I installed the 64 bit version that is posted in the nightly builds (OpenModelica-v1.10.0-dev-158-g96496c0-64bit), and it runs through the corba interface, but Connection editor fails to even open - I get a crash every time.

Thanks for the update!

It looks like the 64 bit version is not being built with the nightly builds. Is that as expected?

This was an error that I got with 1.9.3, I'm now seeing it with 1.10.0 dev9
I get several errors similar to the following:
WebModel.c: In function 'WebModel_eqFunction_10199':
WebModel.c:47889: error: '$PDrivenRoll_5$POut$lB2$rB$PTension' undeclared (first use in this function)
WebModel_06inz.c: In function 'WebModel_eqFunction_5763':
WebModel_06inz.c:70367: error: '$PDrivenRoll_5$POut$lB2$rB$PTension' undeclared (first use in this function)

That solved my problem, thanks! Any estimate when the 64 bit version will be official?

With v1.9.4, I get an error with one of my models that says "too many heap sections", then crashes OMEdit.

Yes, it appears to be corrected in the latest nightly build, thanks! Any idea when this fix will be in an official release?

I try to run a simulation and I get Error Code 3, it says Got number 7443.000000, expected: 7441.000000. This is after it got through translation and compilation phase. This is in version 1.9.4 beta 2. When I try to run the same model in 1.9.3, I get:
WebModel_06inz.c:12220: error: '$P$STATESET0$PA$lB1$c1$rB' undeclared (first use in this function)
WebModel_06inz.c:12220: error: '$P$STATESET0$PA$lB1$c2$rB' undeclared (first use in this function)
WebModel_06inz.c:12220: error: '$P$STATESET0$PA$lB1$c5$rB' undeclared (first use in this function)
WebModel_06inz.c:12220: error: '$P$STATESET0$PA$lB1$c7$rB' undeclared (first use in this function)
WebModel_06inz.c:12220: error: '$P$STATESET0$PA$lB1$c8$rB' undeclared (first use in this function)
WebModel_06inz.c:12220: error: '$P$STATESET0$PA$lB1$c9$rB' undeclared (first use in this function)
WebModel_06inz.c:12220: error: '$P$STATESET0$PA$lB1$c12$rB' undeclared (first use in this function) With many more very similar things before and after. It looks like it is translating, but giving me this error in the compile phase.

I tried creating that directory, and placing the file in that directory, but it still didn't seem to recognize it. The annotation line I have (someone else wrote it for me, so I'm not sure why it was done this way or what it means) is:
  annotation(Include = "#include \"ModelicaRNG.c\"");

This is the annotation in a modelica function that calls it. With that annotation, should the file be in the same folder as the modelica function?

Thanks.

The rest of the error is (with names removed to make it more generic):
In file included from[<LibraryName>.<ModelName>_functions.c:7:
[[<LibraryName>.<ModelName>_includes.h:4:25: error: <ExternalFunctionName>.c: No such file or directory
mingw32-make: *** [<LibraryName>.<ModelName>_functions.o] Error 1
mingw32-make: *** Waiting for unfinished jobs....
Compilation process failed. Exited with code 2.

I'm trying to use an external c function in v1.9.3, and I get an error that the file does not exist. I have the file located in several places within the folder structure, but it still tells me that the file does not exist. I have also verified the spelling, and have tried it with the library as a single file, but nothing seems to get rid of the error. Please advise.

I'm using the latest official release - 1.9.2.

Any plans for a 64-bit Windows version?

Unfortunately, due to the proprietary nature of my work I can't share it directly, but I can tell you that I also got several warnings that it got to the maximum number of iterations without being able to calculate the roots (not sure if the two things are related).

What I was hoping is that someone could tell me what "out of memory" means, since the memory usage shown on the Windows task manager stayed around 5GB, and I have 16GB of RAM available on my computer.

Thanks

I'm using mat format

I was able to get past the issue by increasing the step size (decreasing the number of steps), but I don't understand why I am hitting any kind of limit in the first place, or how I can get past the problem without having to change the step size.

Thanks.

I am having trouble getting the following parameter definition to translate properly:

parameter Real C[4,2] = cat(1, {{0,0}}, transpose({array(B * i for i in 2:4),array(A * B * i for i in 2:4)}));

The error I get is:
Error: Instantiation of array component: C failed because index modification: (C[]= = {{0.0, 0.0}}) is invalid.
    Array component: C[] has more dimensions than binding  = {{0.0, 0.0}}

Any ideas what could cause that?

Thanks!