Installation Question

[Erik Anderson]

Hello PRISMS-PF users,

I was just able to install deal.ii via candi on my machine. But when I follow the instructions to build PRISMS-PF, I get the following error. 

(blah/blah/blah) deal.II-v9.6.0/include/deal.II/matrix_free/fe_evaluation.h:6648:1: note:   candidate expects 1 argument, 3 provided.

I also see several other errors for different programs with the wrong number of arguments provided. Is PRISMS-PF not compatible with the latest version of deal.ii? 

Please advise,

Erik J. Anderson

[Jason Landini]

Hi Erik,

The master branch of prisms-pf is not compatible with deal.II 9.6.0. However, the development branch does if you switch to that branch.

Let me know if you have any other questions,

Jason

--
You received this message because you are subscribed to the Google Groups "PRISMS-PF Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to prisms-pf-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/prisms-pf-users/68d380e8-1b73-4b01-bada-aca26a10371cn%40googlegroups.com.

[Erik Anderson]

Jason,

So, PRISMS-PF said it compiled when I cloned the development branch. But now I'm getting this error.

CMake Error at CMakeLists.txt:9 (find_package):
  Could not find a package configuration file provided by "deal.II"
  (requested version 9.2.0) with any of the following names:

    deal.IIConfig.cmake
    deal.ii-config.cmake

  Add the installation prefix of "deal.II" to CMAKE_PREFIX_PATH or set
  "deal.II_DIR" to a directory containing one of the above files.  If
  "deal.II" provides a separate development package or SDK, be sure it has
  been installed.

Sincerely,

Erik J. Anderson

[Jason Landini]

A few questions,

1. Did you delete CMakeCache.txt before recompiling?

2. Did you specify the path to dealii in your .bashrc file (i.e., export DEAL_II_DIR=/path/to/candi/deal.II-v9.6.0)?

3. What OS are you using?

To view this discussion on the web visit https://groups.google.com/d/msgid/prisms-pf-users/f1bc3e84-e723-4207-9126-db626d7f15edn%40googlegroups.com.

--

Jason Landini

Ph.D. Pre-Candidate, Thornton Group

Materials Science & Engineering

University of Michigan

land...@umich.edu

[Erik Anderson]

1. I did not (I can do that.)

2. I did the second time I tried compiling the AllenCahn example (I'll do that again after removing the CMakeCache)

3. I'm using windows 10 with WSL.

Erik

[Jason Landini]

Ok, thanks!

We are planning on updating the documentation for candi installations with PRISMS-PF soon, so let us know if run into any other issues or if anything is confusing.

[Erik Anderson]

Hi Jason,

So I was able to get the AllenCahn example to run. But when I tried running a few other example applications the same way, I got this error. The AllenCahn example still runs when I recompile it, but for some reason I get this error whenever I try to run another example application.

################################################################################################

mpiexec -n 4 ./main
Using the input parameter file: parameters.prm
Number of constants: 8
Number of post-processing variables: 2

--------------------------------------------------------
An error occurred in line <1058> of file </home/ejanders/dealii-candi/tmp/unpack/deal.II-v9.6.0/source/base/parameter_handler.cc> in function
    std::string dealii::ParameterHandler::get(const std::string&) const
Additional information:
    You can't ask for entry <x> you have not yet declared.

Stacktrace:
-----------
#0  /home/ejanders/dealii-candi/deal.II-v9.6.0/lib/libdeal_II.g.so.9.6.0: dealii::ParameterHandler::get(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const
#1  /home/ejanders/dealii-candi/deal.II-v9.6.0/lib/libdeal_II.g.so.9.6.0: dealii::ParameterHandler::get_double(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const
#2  ./main: userInputParameters<2>::userInputParameters(inputFileReader&, dealii::ParameterHandler&, variableAttributeLoader)
#3  ./main: main
--------------------------------------------------------

Calling MPI_Abort now.
To break execution in a GDB session, execute 'break MPI_Abort' before running. You can also put the following into your ~/.gdbinit:
  set breakpoint pending on
  break MPI_Abort
  set breakpoint pending auto
Abort(255) on node 0 (rank 0 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 255) - process 0

################################################################################################

I haven't found any similar errors like this in the forum, I'm not sure if this is an issue with the developer branch or if I'm doing something wrong with these examples.

Sincerely,

Erik J. Anderson

[David Montiel]

Hello, Erik

What was the example application that failed? Did you modify anything?

The error appears to indicate that you are trying to access a variable that has not being declared.

Best,

David

To view this discussion visit https://groups.google.com/d/msgid/prisms-pf-users/4205b324-e414-40f2-ba53-87cca48d3597n%40googlegroups.com.

[Erik Anderson]

David,

I tried this on the DendriticSolidification and CoupledCahnHilliardAllenCahn applications and got the same error. I don't believe I modified anything, I just compiled the main executable as with the AllenCahn example.

Sincerely,

Erik 

[David Montiel]

@Jason Landini, could you please look into this? I don't have access to an installation of deal.II v9.6 at the moment.

Thanks!

To view this discussion visit https://groups.google.com/d/msgid/prisms-pf-users/1bd7478c-eafe-492b-9429-766c8a6556aen%40googlegroups.com.

[Jason Landini]

I am unable to reproduce this error on the latest version of development.

Can you try pulling the latest version? Alternatively, type git status to see if anything from those applications comes up.

You received this message because you are subscribed to a topic in the Google Groups "PRISMS-PF Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/prisms-pf-users/YZP1xkJGjhk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to prisms-pf-use...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/prisms-pf-users/1bd7478c-eafe-492b-9429-766c8a6556aen%40googlegroups.com.

[Jason Landini]

This may be an issue on the developer branch. What applications are giving this error?

[Erik Anderson]

Jason,

I have tried it with pretty much every application I've tried (I can't remember all of them I've tried but I know I did Coupled AllenCahn, Dendrite solidification, advection, and a couple others) with the exception of Allen Cahn. 

Sincerely,

Erik

[Jason Landini]

Hi Erik,

Can you type git log and let me know what the latest commit was?

Also, have you tried recompiling the whole library by typing "make clean" and deleting "CMakeCache.txt".

To view this discussion visit https://groups.google.com/d/msgid/prisms-pf-users/a39c55cf-5177-4825-8a9f-fcd98bd3c851n%40googlegroups.com.

[Erik Anderson]

I'm seeing Jason Landini, Date Friday October 25th 15:06:19 2024

Erik

[Erik Anderson]

Getting the same error when I try recompiling and deleting CMakeCache.txt

ejanders@:~/phaseField/applications/dendriticSolidification$ mpiexec -n 4 ./main

Using the input parameter file: parameters.prm

Number of constants: 6
Number of post-processing variables: 0

Abort(255) on node 1 (rank 1 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 255) - process 1

Calling MPI_Abort now.
To break execution in a GDB session, execute 'break MPI_Abort' before running. You can also put the following into your ~/.gdbinit:
  set breakpoint pending on
  break MPI_Abort
  set breakpoint pending auto

Abort(255) on node 2 (rank 2 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 255) - process 2

Calling MPI_Abort now.
To break execution in a GDB session, execute 'break MPI_Abort' before running. You can also put the following into your ~/.gdbinit:
  set breakpoint pending on
  break MPI_Abort
  set breakpoint pending auto

Abort(255) on node 3 (rank 3 in comm 0): application called MPI_Abort(MPI_COMM_WORLD, 255) - process 3

[Jason Landini]

There was a short period where all the applications with multiple fields may have been broken (due to that commit).

Can you pull the latest version of development. It should work after that.

To view this discussion visit https://groups.google.com/d/msgid/prisms-pf-users/af692f28-60a3-46ae-8d7d-1d6900831ae2n%40googlegroups.com.

[Erik Anderson]

Jason & David

Just an update. It is working now. I've been able to successfully run a few of the example applications! Thank you for the help.

Sincerely,

Erik

[Jason Landini]

Glad to hear! And sorry for the trouble with the dev branch.

You received this message because you are subscribed to a topic in the Google Groups "PRISMS-PF Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/prisms-pf-users/YZP1xkJGjhk/unsubscribe.
To unsubscribe from this group and all its topics, send an email to prisms-pf-use...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/prisms-pf-users/ed0d6b8e-71f0-4e13-baf8-58891589b6e7n%40googlegroups.com.