Re: [DuMux] Exercice-fractures

[Christoph Grüninger]

Hi Suzon,
adding nonisothermal effects makes the system more non-linear. Hence, 
the Newton solver has a harder time to find a solution.


You can try to:
- Allow more Newton steps
- Reduce the time step size, limit its growth, or just use super tiny steps.
- Reduce the temperature difference, maybe fix the temperature with a 
Dirichlet value everywhere to a single value and use the same value as 
the initial value for the temperature


If you are lucky, this helps to get the simulation running for a couple 
of steps and you can maybe identify areas of nonphysical behavior - 
caused by a bug in your code.


Bye
Christoph


Am 23.04.24 um 12:43 schrieb Suzon Jammes:

Hi all,

I was able to include nonisothermal effects in the fractures setup but I 
am now stuck trying to use the TwoPNCNI  or TwoPNC models instead of the 
TwoPmodel. The code compile but the Newton solver doesn't converge and 
the run stop.


I tried to modify the boundary condition in implementing some injection 
as in the basic exercise but I get the same problem.


What else do I need to modify to use several components in this setup ?

Thank you for your help,

Suzon


On 20/03/2024 13:33, Martin Schneider wrote:

Dear Suzon,

did you change the the model in properties.hh?
If you want to include nonisothermal effects you have to use the 
TwoPNI model instead of the TwoP model.


Best,
Martin

On 19.03.24 17:36, Suzon Jammes wrote:

Dear users,

To learn how to use dumux I am trying to modify a bit the exercise 
about fractures 
(https://git.iws.uni-stuttgart.de/dumux-repositories/dumux-course/-/blob/master/exercises/exercise-fractures). Today I wanted to make it nonisothermal as explained in exercice-basic but I am getting the following error :


error: ‘temperatureIdx’ is not a member of 
‘Dumux::MatrixSubProblem::Indices’ {aka ‘Dumux::TwoPIndices’}


Is there anybody who knows how to solve this error ?

Thank you for your help,

Suzon


___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux





___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux


--
L'enjeu est de bâtir la France de nos enfants,
pas de ressasser la France de notre enfance.
   [Emmanuel Macron, 2022]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Non isothermal coupled exercise

[Christoph Grüninger]

Hi Suzon,
check if the linear system is set up correctly. Reduce the grid size to 
as small as possible (like 1x6) and check if the matrix contains entries 
everywhere you expect it. If non-zero entries are missing in the linear 
system (matrix), you have an indictaion where to look for the error.


Bye
Christoph

--
L'enjeu est de bâtir la France de nos enfants,
pas de ressasser la France de notre enfance.
   [Emmanuel Macron, 2022]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Non isothermal coupled exercise

[Christoph Grüninger]

Hi Suzon,
I did not look into the code, so I might guess wrong.

I suspect that by adding temperature, you got a third primary variable. 
The error points out that you are using a vector with two entries 
(representing two primary variables) at a place where you should use a 
vector with three entries (for all three primary vars).


I hope this helps you figuring out what went wrong.

Bye
Christoph


Am 29.03.24 um 11:14 schrieb Suzon Jammes:

Hi all,

I am now working from the coupled exercise 
(exercises/exercise-coupling-ff-pm/models) with a 2pnc model. I wanted 
to make it non-isothermal so I change TwoPNC by TwoPNCNI and 
NavierStockesNC by NavierStockesNCNI in the properties file but I get 
the following error :


error: conversion from ‘FieldVector<[...],2>’ to non-scalar type 
‘FieldVector<[...],3>’ requested
   263 |  elemVolVars[scvf.insideScvIdx()].extrusionFactor() * 
FluidSystem::molarMass(1) * -1.0 * 86400.0;


I don't understand what does it mean. Does anybody know how to fix this 
problem ?


Thank you

Suzon Jammes


___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux


--
L'enjeu est de bâtir la France de nos enfants,
pas de ressasser la France de notre enfance.
   [Emmanuel Macron, 2022]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Help for the failure of Installation

[Christoph Grüninger]

Hello Junqiang Wang,looks like you have an older installation of Dune and Dumux. This might interfere with checking out the right code. Can you please retry in an empty folder?ByeChristophAm 28.12.2023 um 03:01 schrieb 王俊强（ Junqiang Wang ） :





I encountered an error while using the Python script in the Dumux installation instructions for installation. I am not sure what caused the installation failure. Following the installation failure prompt, I would like to ask for advice on how to resolve it.The attachment is the installation log. Looking forward to your reply




 


Junqiang Wangjunqiangwang_...@163.com


 

___DuMux mailing listDuMux@listserv.uni-stuttgart.dehttps://listserv.uni-stuttgart.de/mailman/listinfo/dumux___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] global cell index for 3D SPgrid with MPI

[Christoph Grüninger]

estions have to do with the dune core code rather than the dumux 
code. So I suppose I should then contact directly the dune mailing list?


Regards,
Mona


*From:* DuMux  on behalf of 
Christoph Grüninger 

*Sent:* Tuesday, October 24, 2023 4:12:28 PM
*To:* dumux@listserv.uni-stuttgart.de
*Subject:* Re: [DuMux] global cell index for 3D SPgrid with MPI
Hi Mona,
Dune 2.6 and DuMuX 3.0 are a bit outdated. Can you reproduce your issue
with the latest stable versions of DuMuX, Dune 2.9 and SPGrid 2.9, too?
I know, easier said then done.
Does this problem occur with YaspGrid and >7 cores?

I am not sure whether you are hitting a SPGrid issues or a DuMuX issue.

Bye
Christoph


Am 24.10.23 um 11:52 schrieb Giraud, Mona:

Dear DuMux team,
I use dumux 3.0 and dune 2.6.
I encountered an issue when computing a problem using 3D SPgrid on ‘too 
many’ cores.
Indeed, it looks like the cells’ ids according to Dune::GlobalIndexSet 
is incorrect when the core/cell ratio is too high.
I set up a small example using an adapted version of 1st exercise of the 
dumux course (see zip folder attached).



##  changes made to 'exercise_basic_2pni_solution' (from 
DuMux 3.0):



In injection2pniproblem.hh:

#include 
[…]
// Set the grid type
template
struct Grid { using type = 
Dune::SPGrid, 3>; };



In exercise_basic_2pni.cc:
// getDofIndices, getPointIndices, getCellIndices
#include 

[...]
int main(int argc, char** argv) try
{
      using namespace Dumux;

      // define the type tag for this problem
      using TypeTag = Properties::TTag::Injection2pNICC;

      // initialize MPI, finalize is done automatically on exit
      const auto& mpiHelper = Dune::MPIHelper::instance(argc, argv);

      // print dumux start message
      if (mpiHelper.rank() == 0)
          DumuxMessage::print(/*firstCall=*/true);

      // parse command line arguments and input file
      Parameters::init(argc, argv);

      // try to create a grid (from the given grid file or the input file)
      GridManager> gridManager;
      gridManager.init();

      
      // run instationary non-linear problem on this grid
      

      // we compute on the leaf grid view
      const auto& leafGridView = gridManager.grid().leafGridView();

      // create the finite volume grid geometry
      using FVGridGeometry = GetPropTypeProperties::FVGridGeometry>;

      using SoilGridType = GetPropType;
using GridView = typename SoilGridType::Traits::LeafGridView;
enum {
dimWorld = GridView::dimensionworld
};
      auto fvGridGeometry = std::make_shared(leafGridView);
      fvGridGeometry->update();

auto cellIdx = 
std::make_shared>(gridManager.grid().leafGridView(), 0);
std::cout<<"getCellCenters rank:"<dimWorld:"<
          for (const auto& e : elements(fvGridGeometry->gridView())) {
              auto p = e.geometry().center();
int gIdx = cellIdx->index(e);//get the cell global index
std::cout<<"gIdx "<          gIdx 6, 0:7.5, 1:7.5, 2:2.5 // != gIdx 6, 0:2.5, 1:7.5, 2:7.5 
obtained when using 1 core

          gIdx 1, 0:2.5, 1:2.5, 2:7.5
          gIdx 5, 0:7.5, 1:2.5, 2:7.5
          gIdx 3, 0:2.5, 1:7.5, 2:7.5
          gIdx 7, 0:7.5, 1:7.5, 2:7.5
getCellCenters rank:2 dimWorld:3
          gIdx 0, 0:2.5, 1:2.5, 2:2.5
          gIdx 4, 0:7.5, 1:2.5, 2:2.5
          gIdx 2, 0:2.5, 1:7.5, 2:2.5
          gIdx 6, 0:7.5, 1:7.5, 2:2.5
          gIdx 1, 0:2.5, 1:2.5, 2:7.5
          gIdx 5, 0:7.5, 1:2.5, 2:7.5
          gIdx 3, 0:2.5, 1:7.5, 2:7.5
          gIdx 7, 0:7.5, 1:7.5, 2:7.5
getCellCenters rank:3 dimWorld:3
          gIdx 0, 0:2.5, 1:2.5, 2:2.5
          gIdx 4, 0:7.5, 1:2.5, 2:2.5
          gIdx 2, 0:2.5, 1:7.5, 2:2.5
          gIdx 6, 0:7.5, 1:7.5, 2:2.5
          gIdx 1, 0:2.5, 1:2.5, 2:7.5
          gIdx 5, 0:7.5, 1:2.5, 2:7.5
          gIdx 3, 0:2.5, 1:7.5, 2:7.5
          gIdx 7, 0:7.5, 1:7.5, 2:7.5

##


As can be seen, when 4 cores are used, the global cell index does not 
follow the 'default' dumux ordering (from the bottom left to the top 
right), although the local cell index still seems to follow it.
The problem seems to disappear when the number of cells is higher. E.g., 
with :


[Grid]
LowerLeft = 0 0 0
UpperRight = 100 100 100
Cells = 5 5 20

Works again for a number of cores <= 7.

Would you know how to fix this issue? Indeed, in our simulation we would 
need to use a higher number of cores (> 7) with a rather small soil grid 
(~ 500 cells).

Many thanks for your help!
Best,
Mona

___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux 
<https://listserv.uni-stuttgart.de/mailman/listinfo/dumux>

Re: [DuMux] global cell index for 3D SPgrid with MPI

[Christoph Grüninger]

Hi Mona,
Dune 2.6 and DuMuX 3.0 are a bit outdated. Can you reproduce your issue 
with the latest stable versions of DuMuX, Dune 2.9 and SPGrid 2.9, too? 
I know, easier said then done.

Does this problem occur with YaspGrid and >7 cores?

I am not sure whether you are hitting a SPGrid issues or a DuMuX issue.

Bye
Christoph


Am 24.10.23 um 11:52 schrieb Giraud, Mona:

Dear DuMux team,
I use dumux 3.0 and dune 2.6.
I encountered an issue when computing a problem using 3D SPgrid on ‘too 
many’ cores.
Indeed, it looks like the cells’ ids according to Dune::GlobalIndexSet 
is incorrect when the core/cell ratio is too high.
I set up a small example using an adapted version of 1st exercise of the 
dumux course (see zip folder attached).



##  changes made to 'exercise_basic_2pni_solution' (from 
DuMux 3.0):



In injection2pniproblem.hh:

#include 
[…]
// Set the grid type
template
struct Grid { using type = 
Dune::SPGrid, 3>; };



In exercise_basic_2pni.cc:
// getDofIndices, getPointIndices, getCellIndices
#include 

[...]
int main(int argc, char** argv) try
{
     using namespace Dumux;

     // define the type tag for this problem
     using TypeTag = Properties::TTag::Injection2pNICC;

     // initialize MPI, finalize is done automatically on exit
     const auto& mpiHelper = Dune::MPIHelper::instance(argc, argv);

     // print dumux start message
     if (mpiHelper.rank() == 0)
         DumuxMessage::print(/*firstCall=*/true);

     // parse command line arguments and input file
     Parameters::init(argc, argv);

     // try to create a grid (from the given grid file or the input file)
     GridManager> gridManager;
     gridManager.init();

     
     // run instationary non-linear problem on this grid
     

     // we compute on the leaf grid view
     const auto& leafGridView = gridManager.grid().leafGridView();

     // create the finite volume grid geometry
     using FVGridGeometry = GetPropTypeProperties::FVGridGeometry>;

     using SoilGridType = GetPropType;
using GridView = typename SoilGridType::Traits::LeafGridView;
enum {
dimWorld = GridView::dimensionworld
};
     auto fvGridGeometry = std::make_shared(leafGridView);
     fvGridGeometry->update();

auto cellIdx = 
std::make_shared>(gridManager.grid().leafGridView(), 0);
std::cout<<"getCellCenters rank:"gridView())) {
             auto p = e.geometry().center();
int gIdx = cellIdx->index(e);//get the cell global index
std::cout<<"gIdx "<         gIdx 6, 0:7.5, 1:7.5, 2:2.5 // != gIdx 6, 0:2.5, 1:7.5, 2:7.5 
obtained when using 1 core

         gIdx 1, 0:2.5, 1:2.5, 2:7.5
         gIdx 5, 0:7.5, 1:2.5, 2:7.5
         gIdx 3, 0:2.5, 1:7.5, 2:7.5
         gIdx 7, 0:7.5, 1:7.5, 2:7.5
getCellCenters rank:2 dimWorld:3
         gIdx 0, 0:2.5, 1:2.5, 2:2.5
         gIdx 4, 0:7.5, 1:2.5, 2:2.5
         gIdx 2, 0:2.5, 1:7.5, 2:2.5
         gIdx 6, 0:7.5, 1:7.5, 2:2.5
         gIdx 1, 0:2.5, 1:2.5, 2:7.5
         gIdx 5, 0:7.5, 1:2.5, 2:7.5
         gIdx 3, 0:2.5, 1:7.5, 2:7.5
         gIdx 7, 0:7.5, 1:7.5, 2:7.5
getCellCenters rank:3 dimWorld:3
         gIdx 0, 0:2.5, 1:2.5, 2:2.5
         gIdx 4, 0:7.5, 1:2.5, 2:2.5
         gIdx 2, 0:2.5, 1:7.5, 2:2.5
         gIdx 6, 0:7.5, 1:7.5, 2:2.5
         gIdx 1, 0:2.5, 1:2.5, 2:7.5
         gIdx 5, 0:7.5, 1:2.5, 2:7.5
         gIdx 3, 0:2.5, 1:7.5, 2:7.5
         gIdx 7, 0:7.5, 1:7.5, 2:7.5

##


As can be seen, when 4 cores are used, the global cell index does not 
follow the 'default' dumux ordering (from the bottom left to the top 
right), although the local cell index still seems to follow it.
The problem seems to disappear when the number of cells is higher. E.g., 
with :


[Grid]
LowerLeft = 0 0 0
UpperRight = 100 100 100
Cells = 5 5 20

Works again for a number of cores <= 7.

Would you know how to fix this issue? Indeed, in our simulation we would 
need to use a higher number of cores (> 7) with a rather small soil grid 
(~ 500 cells).

Many thanks for your help!
Best,
Mona

___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Problems installing Dumux

[Christoph Grüninger]

Hello Xavier,

you don't have all required dependencies installed to have Python 
enabled. I assume starting working with DuMuX is okay for you.


I don't know why, although Python is not found, DuMuX tried to run 
Python-related commands - which fails.


Quick hack to get you a working DuMuX installation:
Write comments (#) in front of every line in
dumux/dumux/python/dumux/assembly/CMakeLists.txt
(relative to your installdumux.py).

We have to figure out why DuMuX tries to use Python and fix the issue.

Bye
Christoph


Am 27.07.23 um 10:24 schrieb Xavier Tort Farrés:

Dear Sir or Madam,

I am trying to install Dumux, and I already had some warning messages 
that I think I could fix. Even though, at step (3/3), which corresponds 
to /Configure and build dune modules and dumux using dunecontrol /an 
error message appear, which is:


(Error) The command ['./dune-common/bin/dunecontrol', 
'--opts=dumux/cmake.opts', 'all'] returned with non-zero exit code


I am new to Linux, so I do not have experience on solving errors like 
this. This is why I made a search on the Internet but I did not found a 
clear answer, so I am writing to you to see if you could help me as it 
will be very useful for me.


As mentioned in the error, I also attached the file 'installdumux.log'.

Thank you in advance for your help.

Best regards,
--

*Xavier Tort Farrés**
Research Eng. - Structural Mechanics Department.***
CIMNE - Building C1, Floor 2, Office C6 - Campus Nord UPC
C/ Gran Capità, S/N
08034 Barcelona, Spain
Tel. - +34 93 401 73 99
*www.cimne.com* 

*CIMNE Severo Ochoa

*


___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux


--
L'enjeu est de bâtir la France de nos enfants,
pas de ressasser la France de notre enfance.
   [Emmanuel Macron, 2022]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] About DuMux Installation

[Christoph Grüninger]

Hello Yigun,

the attached log file is not very helpful, it lacks quite some 
information. This is not your fault, I improved the script. If you want, 
you can use the attached version.


Part of the log file is the following line:
> dumux/cmake.opts: file not found
This indicates, that from the folder where dunecontrol is executed, no 
dumux/cmake.opts was found. I don't know how this can happen. Can you 
check whether the file dumux/dumux/cmake.opts exists relative to your 
installdumux.py file?


Further, you can execute in the dumux forder (not dumux/dumux) the 
following command and share the output:

./dune-common/bin/dunecontrol --opts=dumux/cmake.opts all

Bye
Christoph


Am 20.07.23 um 06:03 schrieb Yiqun Ma:

Dear Sir/Madam,

I am a newbie to DuMux and I am getting this error while installing 
using the script in 
https://dumux.org/docs/doxygen/master/installation.html 
. What should I 
do to fix it please?


All the best,

Yiqun#!/usr/bin/env python3
# SPDX-FileCopyrightInfo: Copyright © DuMux Project contributors, see AUTHORS.md in root folder
# SPDX-License-Identifier: GPL-3.0-or-later


"""
One click install script for dumux
"""
import os
import sys
import argparse
import subprocess
import textwrap
from shutil import which


class _Version:
def __init__(self, version: str) -> None:
self._version = [int(v) for v in version.strip(" ").strip("\n").split(".")]

def __lt__(self, other) -> bool:
for versionSelf, versionOther in zip(self._version, other._version):
if versionSelf < versionOther:
return True
if versionSelf > versionOther:
return False
return False


parser = argparse.ArgumentParser(
prog="installdumux",
usage="./installdumux.py [OPTIONS]",
description="This script downloads and compiles the latest release of Dumux.",
)
# Optional arguments
parser.add_argument("--dune-version", default="2.9", help="Dune version to be checked out.")
parser.add_argument("--dumux-version", default="3.7", help="Dumux version to be checked out.")
args = vars(parser.parse_args())

duneBranch = (
args["dune_version"] if args["dune_version"] == "master" else "releases/" + args["dune_version"]
)
dumuxBranch = (
args["dumux_version"]
if args["dumux_version"] == "master"
else "releases/" + args["dumux_version"]
)


def showMessage(message):
"""Pretty print message"""
print("*" * 120)
print(message)
print("*" * 120)


def checkCppVersion():
"""Check compiler version"""
requiredversion = "9.3"
result = subprocess.check_output(["g++", "-dumpversion"]).decode().strip()
if _Version(result) < _Version(requiredversion):
print("-- An error occurred while checking for prerequistes.")
raise Exception(
f"g++ greater than or equal to {requiredversion} "
"is required for dumux releases >=3.2!"
)


def runCommand(command, workdir="."):
"""Run command with error checking"""
with open("../installdumux.log", "a") as log:
print("> ", " ".join(command), file=log)
with subprocess.Popen(
command,
stdout=log,
stderr=log,
universal_newlines=True,
cwd=workdir,
) as popen:
returnCode = popen.wait()
if returnCode:
message = textwrap.dedent(
f"""\

(Error) The command {command} returned with non-zero exit code
If you can't fix the problem yourself consider reporting your issue
on the mailing list (dumux@listserv.uni-stuttgart.de) and
attach the file 'installdumux.log'
"""
)
showMessage(message)
print(message, file=log)
sys.exit(1)


def gitClone(url, branch=None):
"""Clone git repo"""
clone = ["git", "clone"]
if branch:
clone += ["-b", branch]
runCommand(command=[*clone, url])


def gitSetBranch(folder, branch):
"""Checkout specific git branch"""
checkout = ["git", "checkout", branch]
runCommand(command=checkout, workdir=folder)


# clear the log file
with open("installdumux.log", "w") as _:
pass

log = open('installdumux.log', 'a')

#
#
# (1/3) Check some prerequistes
#
#
programs = ["git", "gcc", "g++", "cmake", "pkg-config"]
showMessage("(1/3) Checking all prerequistes: " + " ".join(programs) + "...")
print("(1/3) Checking all prerequistes: " + " ".join(programs) + "...", file=log)

# check some prerequistes
for program in programs:
if which(program) is None:
print("-- An 

Re: [DuMux] REQUEST FOR SUPPORT TO INSTALL DUMUX ON UBUNTU SHELL ON WINDOWS 10 PC

[Christoph Grüninger]

Hi Yussif,
difficult to tell what is going wrong. Your log only contains 
information regarding the Git checkout. The posted error tells me, that 
dunecontrol failed, but the actual error was consumed.

(I assume you are using installdumux.py)

Can you please run in some WSL terminal
./dune-common/bin/dunecontrol --opts=dumux/cmake.opts all
and report the output? Run the command in the directory that contains 
dune-common.


Bye
Christoph

Am 21.06.23 um 17:25 schrieb Yussif Mustapha:

Hello!
  My name is Yussif Mustapha. I am currently a Petroleum Engineering 
MS student at Middle East Technical University, Ankara, Turkiye, and I 
am working on Geomechanics and fluid flow as part of my thesis.


I had to follow the hard way of running DuMux on an UBUNTU shell on my 
Windows PC, and this is the error (kindly find it below) I have received 
while carefully following the steps. It talks about an 
"installdumux.log" file to be attached. (Kindly find it attached).


I hereby request that you kindly provide me with the needed guidance in 
order to be able to install and run the dumux simulator on my PC.


I count on your much anticipated cooperation and support.

Thank you!
Yussif.
Tel: +905525633090

(Error) The command ['./dune-common/bin/dunecontrol', 
'--opts=dumux/cmake.opts', 'all'] returned with non-zero exit code

If you can't fix the problem yourself consider reporting your issue
on the mailing list (dumux@listserv.uni-stuttgart.de 
) and

attach the file 'installdumux.log'

___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux


--
L'enjeu est de bâtir la France de nos enfants,
pas de ressasser la France de notre enfance.
   [Emmanuel Macron, 2022]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Permeability interface problem

[Christoph Grüninger]

Hi Kenza,
how do you get the data for your red line and how do you plot them?

- I assume that the permeability is constant for a complete cell. Is 
this correct?
- If you get the value for the box center, you should plot the value for 
the whole cell with the value from the center. Then you get the sharp 
interfaces in your plot.


With other words, the unexpected plot results from a wrong visualization.

Bye
Christoph

Am 14.03.23 um 14:26 schrieb kenza bouznari:

Dear Dumux users,

My simulations show a progressive permeability interface between two 
different materials instead of a sharp transition from one material to 
another. Here's the permeability profile I obtain & the assumed 
interfaces between the different layers in my conceptual model:

image.png
I am using test/2p2c/injection, the box method. The problem persists 
even with a local refinement of the grid.


Do you have any suggestions for possible ways how to capture the sharp 
interface between two different materials?


Thank you in advance for your help.

Kenza

___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux


--
L'enjeu est de bâtir la France de nos enfants,
pas de ressasser la France de notre enfance.
   [Emmanuel Macron, 2022]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] [DuMuX installation] Unable to install DuMuX on Ubuntu

[Christoph Grüninger]

Salut Timothé,
the output looks like you modified something around your MPI (installed, 
uninstalled, whatever). Remove the build-cmake dir from dune-common and 
try again. Then CMake is going to re-detect MPI and it should work fine.


Hope this helps
Christoph

--
L'enjeu est de bâtir la France de nos enfants,
pas de ressasser la France de notre enfance.
   [Emmanuel Macron, 2022]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] DuMuX 3.6 Release Schedule and Important Information

[Christoph Grüninger]

Hi Mathis,
I had a look at the current state of master, which will eventually 
become DuMuX 3.6.

- I found some warnings and cleanups, see !3276 [1]
- With ALUGrid master I get a couple of warnings, I hope they will fix 
them in a not so distant future.
- There are a ton of warnings for couplingmanager.hh and 
NavierStokesProblem. I hope you (or someone else) will the code to the 
new way. It makes testing harder when the compiler output is flooded by 
the repeated warnings. Further, this will make the removal of the 
deprecated code harder, so it should be remove prior to 3.6.


Hope this helps,
Christoph


[1] 
https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/3276



--
L'enjeu est de bâtir la France de nos enfants,
pas de ressasser la France de notre enfance.
   [Emmanuel Macron, 2022]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Oil-gas compositional simulation in DuMux

[Christoph Grüninger]

Hi Dimitry,
I second what the others say. Please consider publishing your code as a 
merge request. DuMuX and its contributors have proven to be able to 
integrate such codes. I am willing to help, too.


Bye
Christoph

Am 20.03.22 um 02:37 schrieb Edscott Wilson:
Yes, Dmitry, please do consider Bernd's suggestions. It would be 
appreciated by more than one.


Best regards


--
Because apparently, even the invisible hand [of the market] doesn’t
want to pick beans. -- Stephen Colbert to US Congress, 2011
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Installation issues

[Christoph Grüninger]

Hi Vuong Van,
difficult to tell, just from the error message. Can you share a diff on 
your Dumux folder or the CMakeLists.txt?


Bye
Christoph

--
Because apparently, even the invisible hand [of the market] doesn’t
want to pick beans. -- Stephen Colbert to US Congress, 2011
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Installation issues

[Christoph Grüninger]

Hi Vuong Van Pham,

the very first step seems to be failing. I have never seen such a 
problem before.


Can you try to fetch the Dune core libraries manually? The first command 
would be:

git clone https://gitlab.dune-project.org/core/dune-common.git

Kind regards,
Christoph

Am 07.03.22 um 04:56 schrieb Pham, Vuong Van:

To someone who may concern,

My name is Vuong Van Pham, and I am a graduate research assistant who is 
interested in using DuMuX for academic use. While tempting to install 
DuMuX following the guideline, I scoped with the issue (detailed in the 
attached file).


Please respond to this email with further instruction to resolve this issue.

/Best regards,/

*Vuong Van Pham**//*

Graduate Research Assistant (GRA)

Department of Chemical and Petroleum Engineering (CPE), University of Kansas

Email: vuongvanp...@ku.edu 

Phone: +1-(785)-979-2664

*From:* Pham, Vuong Van
*Sent:* 23 Tháng Hai 2022 4:11 CH
*To:* dumux@listserv.uni-stuttgart.de
*Subject:* Installation issues

To someone who may concern,

My name is Vuong Van Pham, and I am a graduate research assistant who is 
interested in using DuMuX for academic use. While tempting to install 
DuMuX following the guideline, I scoped with the issue (detailed in the 
attached file).


Please respond to this email with further instruction to resolve this 
issue.


/Best regards,/

*Vuong Van Pham**//*

Graduate Research Assistant (GRA)

Department of Chemical and Petroleum Engineering (CPE), University of Kansas

Email: vuongvanp...@ku.edu 

Phone: +1-(785)-979-2664


___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux


--
Because apparently, even the invisible hand [of the market] doesn’t
want to pick beans. -- Stephen Colbert to US Congress, 2011
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Integration of boundary flux for upscaling: "errors"

[Christoph Grüninger]

Hi Ana,
I am not sure whether you are right that 1e-14 to 1e-16 should not be 
considered as zero. It is the expected precision of a double. In 
general, you cannot expect it to be better, just because in the regular 
case it by luck is actually better.


Maybe you can run your simulation with quad precision. That might help 
to get more knowledge regarding the limits of double precision.


Bye
Christoph


Am 17.11.21 um 11:31 schrieb Ana Carolina Loyola:

Hello,

I have been working on the upscaling of the 2D permeability tensor of 
fractured media with the Box Method using the multidomain module of 
Dumux 3.2. The code attached has worked well when compared to the 
analytical solutions of perpendicular and equally spaced fractures.
I apply linear pressure boundary conditions and integrate flow at the 
boundaries using the following equation*

image.png
And for that, I created a "boundary flux" function (at main.cc), which 
calls the computeFlux function for all the faces that are located at the 
boundary of the domain.


The reason I send this message is that I have noticed some precision 
errors that concern me when running another simple test case (one 
horizontal and non-persistent fracture, meshes are attached).  It is 
expected that I have kxy and kyx equal to 0, which seems to work fine 
when I use a symmetric mesh (signalized with -sym in the .msh files), 
since I get kxy and kyx in the order of < 1e-30 and net flux balances in 
the order of <1e-28.


But when I use non-symmetric meshes I get kxy in the order of 1e-17, 
which can not be considered to be 0, considering the matrix 
permeability, and boundary fluxes summation as high as 1e-14 when I 
apply the x gradient. The "error" does not seem to improve when 
refining the mesh.


So what I would like to ask here is if I am using the appropriate 
functions to integrate flow and if you would have any 
implementation-related suggestions to get past this issue.


I am sorry for the long text; this may be more of a theoretical issue 
than a code-related one, but I thought it was worth the shot just to see 
if I am missing something.


*Equation used by Pouya and Fouché (2009)
Pouya, Amade & Fouché, Olivier. (2009). Permeability of 3D discontinuity 
networks: New tensors from boundary-conditioned homogenization. Advances 
in Water Resources. 32. 303-314. 10.1016/j.advwatres.2008.08.004.


Thanks

Ana Loyola

___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux



--
Als wär es nix, leb' ich von [IT] und mach' nur, was ich lieb'
Lebe wie im Paradies, womit hab' ich das verdient?
Die Wahrheit ist: Hab' ich nicht, ich bin nur reicher beschenkt
Als jemand in einem armen Land mit dem gleichen Talent
[frei nach Tua von Die Orsons - Oioioiropa]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Unknown CMake command "add_dune_alugrid_flags"

[Christoph Grüninger]

Hi Ranjeet,
this is strange, because add_dumux_test is not part of DuMuX 3.2. It was 
removed after DuMuX 3.1.


As the error message is confusing my, I have to more general questions:
- Did you delete your build directory (by default dumux/build-cmake) and 
rerun dunecontrol?
- Have you updated the Dune modules accordingly? DuMuX 3.2 required 2.6, 
DuMuX 3.3 requires Dune 2.7.


Kind regards,
Christoph

Am 17.11.21 um 09:35 schrieb Ranjeet kumar:

Hi

Currently, I am using dumux3.2 and it works fine. I am trying to update 
it to dumux3.4 but I am getting an error:


|CMake Error at cmake/modules/DumuxTestMacros.cmake:63 
(add_dune_alugrid_flags): Unknown CMake command 
"add_dune_alugrid_flags". Call Stack (most recent call first): 
cmake/modules/DumuxTestMacros.cmake:41 (add_dumux_all_flags) 
test/common/generalproblem/CMakeLists.txt:3 (add_dumux_test) -- 
Configuring incomplete, errors occurred! See also 
"/home/ranjeet/dumuxProjects/dumux3.4/dumux/build-cmake/CMakeFiles/CMakeOutput.log". 
See also 
"/home/ranjeet/dumuxProjects/dumux3.4/dumux/build-cmake/CMakeFiles/CMakeError.log". 
--- Failed to build dumux ---


|

|Regards,
|

|Ranjeet
|


___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux



--
Als wär es nix, leb' ich von [IT] und mach' nur, was ich lieb'
Lebe wie im Paradies, womit hab' ich das verdient?
Die Wahrheit ist: Hab' ich nicht, ich bin nur reicher beschenkt
Als jemand in einem armen Land mit dem gleichen Talent
[frei nach Tua von Die Orsons - Oioioiropa]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] How to setup dumux in an IDE?

[Christoph Grüninger]

Hi Siroos,
you have to run dunecontrol for all your modules first. Then you can use 
an IDE for DuMuX and it should be able to build a specific target.


If you use clangd, even the auto-completion should kind of work. I have 
never tried it myself, but I read about it.


Bye
Christoph


Am 04.10.21 um 08:33 schrieb Siroos Azizmohammadi:

Dear Christoph,

Thank you for your reply and the instructions.
As you mentioned, the dependencies to Dune modules cause problems when 
trying to setup Dumux with IDEs. I have tried Qt with a similar 
procedure you explained but FindPkgConfig.cmake cannot find the Dune 
packages and not start the Dune project even if Qt6 fully supports CMake.


Best regards - Siroos

___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] How to setup dumux in an IDE?

[Christoph Grüninger]

Hi Siroos,

you could use any IDE. Usually they have difficulties with DuMuX, 
because they don't understand the dependencies to the Dune modules.


You could build Dune and DuMuX using dunecontrol, and open the DuMuX 
source folder within your IDE. If your IDE supports CMake, it should be 
able to handle your build directory and it should be able to use the 
right target (make, ninja, msbuild, ...).
This will give you most of the advantages of an IDE, without full 
assistance of Dune classes and functions. In general, IDEs struggle with 
template meta programming and DuMuX' property system.


Bye
Christoph

Am 30.09.21 um 12:37 schrieb Siroos Azizmohammadi:

Hello,

As I am quite new in dumux and Linux too, I was wondering if there are 
any instructions to setup/compile dumux in Qt, Eclipse, or any other IDEs.


Best regards,

_

Siroos Azizmohammadi , PhD



--
Als wär es nix, leb' ich von [IT] und mach' nur, was ich lieb'
Lebe wie im Paradies, womit hab' ich das verdient?
Die Wahrheit ist: Hab' ich nicht, ich bin nur reicher beschenkt
Als jemand in einem armen Land mit dem gleichen Talent
[frei nach Tua von Die Orsons - Oioioiropa]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] PostProcessing

[Christoph Grüninger]

Hi Kenza,
it depend on what data you have (2d or 3d?) and what you need to 
visualize. In any case, I would not include your observed data into 
DuMuX. Instead, you can either print out your simulation data or extract 
them from your Vtk files. Then you can plot them using TikZ, Gnuplot or 
even Excel. Or you can load your observed data into ParaView (or 
whatever tool you use to visualize Vtk files) and use its capabilities 
to show both datasets, calculate differences and so on.


Bye
Christoph


Am 17.09.21 um 19:43 schrieb kenza bouznari:

Hello,

I am not very used to C++ code and I continue my learning curve on 
Dumux! But I want to add some lines of code for Post Processing of 
simulation results. How can we overlay observed data and simulated 
results for some parameters (like fractions).



I am using the 2p2c/chemicalnonequilibrium model of dumux3.4 version.

I want to know if there are any existing tests (in dumux/test directory 
or in dumux-pub repository) that do that or can help to show how it can 
be done?



Thank you all,


KB


--
Als wär es nix, leb' ich von [IT] und mach' nur, was ich lieb'
Lebe wie im Paradies, womit hab' ich das verdient?
Die Wahrheit ist: Hab' ich nicht, ich bin nur reicher beschenkt
Als jemand in einem armen Land mit dem gleichen Talent
[frei nach Tua von Die Orsons - Oioioiropa]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] [DuMuX] DuMuX 3.4 Release: Hard Feature Freeze

[Christoph Grüninger]

Hi Hanchuan,
I tested the master branch with CMake 3.20.1, GCC 11 and Clang 12
respectively. Besides #1029, Clang 12 compilation failure for
test_mpfa2p.cc, it looks good to me.

Have a fine release!

Bye
Christoph


Am 04.05.21 um 16:11 schrieb Wu, Hanchuan:
> Dear DuMuxers,
> 
> 
> Just a quick reminder that tomorrow (*05.05.2021*) we will start
> our "hard freeze" for Dumux release 3.4, which means as of tomorrow:
> 
> 
> Normally only bug fixes and documentation issues can be added as new
> features.
> 
> 
> In case there is any urgent issue, please inform me and the change
> itself must be announced
> in 
> Issue#1009 >
> with a clear description and why it is necessary.
> 
> 
> I have assigned a reviewer to each merge request, please double check
> all the changes. All the open issues and MRs should be addressed this
> week, such that we can test them in the next week.
> 
> 
> Have a nice day!
> 
> 
> Best wishes,
> 
> Hanchuan
> 
> 
> ___
> DuMux mailing list
> DuMux@listserv.uni-stuttgart.de
> https://listserv.uni-stuttgart.de/mailman/listinfo/dumux
> 

-- 
Als wär es nix, leb' ich von [IT] und mach' nur, was ich lieb'
Lebe wie im Paradies, womit hab' ich das verdient?
Die Wahrheit ist: Hab' ich nicht, ich bin nur reicher beschenkt
Als jemand in einem armen Land mit dem gleichen Talent
[frei nach Tua von Die Orsons - Oioioiropa]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Importing an Eclipse Grid

[Christoph Grüninger]

Hi Mahmoud Atef,
looks like a segmentation fault. These are nasty to spot and it is
almost impossible for us to help you.

There are tools to help you finding your mistake, like Valgrind, GDB, or
AddressSanitizer. They often indicate what goes wrong, but you have to
know how to interpret their output.

Good luck!
Christoph
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Error building a new module (dumux-trial)

[Christoph Grüninger]

Hi Mahmoud,

I think it is very ambicious to create a new module without having at
least a little bit of knwolege about programming in general and DuMuX /
Dune in particular.

My suggestion:
- Postpone creating your own module.
- Remove all optional Dune modules.
- Concentrate on using DuMuX: Run some models, modify parameters (input
files), get results visualized with ParaView.
- Add third party libraries or optional Dune modules and start using them.
- Make some minor modifications in the code like changing boundary
conditions, replace van-Genuchten model so something else or vise versa.
- Then you can think about creating your own module.

Further, ask your adviser or other people involved in programming from
your department / chair how to proceed and what you can achieve for your
thesis.

Hope this helps,
Christoph

-- 
Als wär es nix, leb' ich von [IT] und mach' nur, was ich lieb'
Lebe wie im Paradies, womit hab' ich das verdient?
Die Wahrheit ist: Hab' ich nicht, ich bin nur reicher beschenkt
Als jemand in einem armen Land mit dem gleichen Talent
[frei nach Tua von Die Orsons - Oioioiropa]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMux] Recommendation: Dune book

[Christoph Grüninger]

Dear DuMuX users!

I want to recommend Oliver Sander's new Dune book [1] warmly to you! It
was released last December and it covers the Dune core modules,
dune-typetree, -functions, and PDELab in version 2.7. It is a great
source to get started with advanced Dune features like grid refinement,
various aspects of parallelism, and the ISTL solvers like AMG.
DuMuX users who needs to get their hands dirty with such Dune things
will profit from reading the according sections!

Bye
Christoph

[1] Oliver Sander. DUNE — The Distributed and Unified Numerics
Environment. ISBN 978-3-030-59701-6. Springer, 2020.
https://www.springer.com/gp/book/9783030597016
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Vapor pressure of brine

[Christoph Grüninger]

Hi Mohammad,
I think it is this document
https://elib.uni-stuttgart.de/handle/11682/8996

Someone should add the DOI to DUMuX' documentation.

Bye
Christoph

> Am 04.11.2020 um 06:58 schrieb MOHAMMADREZA KARAMI :
> 
> Hi,
> In ../dumux/material/components/brine.hh ,reference of our brine vapor 
> pressure is expressed "simplified version of Eq 2.29 in Vishal Jambhekar's 
> Promo".
> but I can't really find this article.
> Can anyone help me to find this article?
> 
> 
> 
> best regard,
> Mohammad
> -- 
> This email was Anti Virus checked by  Security Gateway.
> ___
> DuMux mailing list
> DuMux@listserv.uni-stuttgart.de
> https://listserv.uni-stuttgart.de/mailman/listinfo/dumux
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Dumux 3.3 Release: Hard Feature Freeze

[Christoph Grüninger]

Hi Kilian,
I gave it a try (with GCC 10 and Clang 11, CMake 3.18 and 3.19rc1, UG
and AluGrid, Dune master and Dune 2.7.1-rc1) and it compiled so far.
I found two minor glitches, both are already resolved in mater.

Have another great release!
Christoph

-- 
Als wär es nix, leb' ich von [IT] und mach' nur, was ich lieb'
Lebe wie im Paradies, womit hab' ich das verdient?
Die Wahrheit ist: Hab' ich nicht, ich bin nur reicher beschenkt
Als jemand in einem armen Land mit dem gleichen Talent
[frei nach Tua von Die Orsons - Oioioiropa]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Dumux and Eclipse - errors which are not errors

[Christoph Grüninger]

Hi Markus,
Kate is a plain editor on steroids, no full fledged IDE. It does not
offer any code analysis.

Probably KDevelop will understand DuMuX code the best, as it includes
clangd. I never used it myself, but it looks promising.

Bye
Christoph

-- 
Mit Kurzschlafwettkampfbestzeit
[Tua von Die Orsons - Sowas von egal]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Dumux and Eclipse - errors which are not errors

[Christoph Grüninger]

Hi Markus,
probably CDT's parser is not good enough to work all his way through the
TypeTag magic DuMuX piles up. There are no additional files you could
include. With Eclipse, I see no way to achieve more.

Clangd could be useful, but as far as I know it is not supported in
Eclipse's CDT. The idea is already at least three years old:
https://www.eclipse.org/community/eclipse_newsletter/2017/april/article5.php

Bye
Christoph

-- 
Mit Kurzschlafwettkampfbestzeit
[Tua von Die Orsons - Sowas von egal]
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMux] Dumux and Eclipse - errors which are not errors

[Christoph Grüninger]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

Hi Markus,
you have to add the config.h file from your DuMuX build directory to
the headers / includes. Out of my head, I don't know how to do that.
But there the missing macros are defined.
By default you can find the file at dumux/build-cmake/config.h

Bye
Christoph

- -- 
Mit Kurzschlafwettkampfbestzeit
[Tua von Die Orsons - Sowas von egal]
-BEGIN PGP SIGNATURE-
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=AHao
-END PGP SIGNATURE-
___
DuMux mailing list
DuMux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Dumux 3.2 Release Candidate

[Christoph Grüninger]

Hi DuMuX,
for the release candidate of DuMuX 3.2 I faced a compiler crash (ICE)
with GCC 10.0 (pre 10.1, first release of GCC 10 series). I don't know
whether you did something or plan to do about it. But I am happy to
share, that the problem got fixed and the fix will be part of GCC 11 and
GCC 10.2 (next major and next series 10 release).

To dignify what they did: A GCC developer reduced a pre-processes DuMuX
file, i.e., with all Dune and C++ includes included from a 830 KiB
compressed source code to less then 100 lines. Another developer spotted
the problematic constellation and provided a fix.

For more details, see
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94937

Bye
Christoph

___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] axisymmetric boundary condition

[Christoph Grüninger]

Hi Timo,
I think Mohammad asks for periodic boundary conditions. Imagine a
turbine with 12 blades. You don't want to simulate the full 360° but
limit the simulation domain to 3 blades (90° piece of circle) or even a
single blade (30° piece of circle).
Neumann no-flow is wrong, you can have inflow and outflow but one side
must match the other.

@Mohammad:
It should work with ALUGrid, which support periodic boundary conditions.
Check its documentation or search the source code for indications on how
to do it. UGGrid should work, too, but I am not sure.
Be aware that periodic boundaries / periodicy is not part of Dune's grid
interface. ALUGrid, UGGrid and YaspGrid all support it, but they do it
different. With yaspgrid you cannot get the geometry.

You will get better answers if you ask on the Dune mailing list.

Bye
Christoph

-- 
Mit Kurzschlafwettkampfbestzeit
[Tua von Die Orsons - Sowas von egal]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Dumux 3.2 Release Candidate

[Christoph Grüninger]

Hi DuMuX,
just for the record, I filed my issue as GCC 94937.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94937

Bye
Christoph

-- 
Mit Kurzschlafwettkampfbestzeit
[Tua von Die Orsons - Sowas von egal]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Dumux 3.2 Release Candidate

[Christoph Grüninger]

Hi Ned,

I gave the current master a try, which should be close enough to the RC.
I found a pedantic warning, which I don't see for DuMuX 3.2, cf. !2110.

What bugs me more, is an ICE with some pre-release version of GCC 10.
ICE is a compiler crash, so it would impact every DuMuX user with GCC
10. As my GCC version is two week old, I don't know whether it is
already fixed or not. I have not yet filed a bug. I hope I will be able
to file one tonight. If someone could confirm this problem, it would be
helpful.
More details regarding the ICE see below.

/home/gruenich/dune/complete/dumux/dumux/flux/cctpfa/fickslaw.hh:188:13:
internal compiler error: in cp_get_fndecl_from_callee, at cp/cvt.c:1000
  188 | if constexpr (FluidSystem::isTracerFluidSystem())
  | ^~

Bye
Christoph

-- 
Mit Kurzschlafwettkampfbestzeit
[Tua von Die Orsons - Sowas von egal]


> g++-10 --version
g++-10 (SUSE Linux) 10.0.1 20200424 (experimental) [revision
3685c5adf5c0b30268cb8f95c89e4c4a3240ae6f]
Copyright (C) 2020 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is
NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
PURPOSE.


[ 33%] Building CXX object
test/multidomain/boundary/stokesdarcy/1p2c_1p2c/CMakeFiles/test_md_boundary_darcy1p2c_stokes1p2c.dir/main.cc.o
In file included from
/home/gruenich/dune/complete/dumux/dumux/flux/fickslaw.hh:48,
 from
/home/gruenich/dune/complete/dumux/dumux/porousmediumflow/properties.hh:39,
 from
/home/gruenich/dune/complete/dumux/dumux/porousmediumflow/1pnc/model.hh:65,
 from
/home/gruenich/dune/complete/dumux/test/multidomain/boundary/stokesdarcy/1p2c_1p2c/problem_darcy.hh:32,
 from
/home/gruenich/dune/complete/dumux/test/multidomain/boundary/stokesdarcy/1p2c_1p2c/main.cc:54:
/home/gruenich/dune/complete/dumux/dumux/flux/cctpfa/fickslaw.hh: In
instantiation of ‘static Dumux::FicksLawImplementation::Scalar
Dumux::FicksLawImplementation::calculateTransmissibility(const Problem&, const
Element&, const FVElementGeometry&, const ElementVolumeVariables&, const
SubControlVolumeFace&, int, int) [with TypeTag =
Dumux::Properties::TTag::DarcyOnePTwoC;
Dumux::ReferenceSystemFormulation referenceSystem =
Dumux::ReferenceSystemFormulation::massAveraged;
Dumux::FicksLawImplementation::Scalar = double;
Dumux::FicksLawImplementation::Problem =
Dumux::DarcySubProblem;
Dumux::FicksLawImplementation::Element =
Dune::Entity<0, 2, const Dune::YaspGrid<2>, Dune::YaspEntity>;
Dumux::FicksLawImplementation::FVElementGeometry
=
Dumux::CCTpfaFVElementGeometry > >, false,
Dumux::CCTpfaDefaultGridGeometryTraits > >,
Dumux::DefaultMapperTraits > >,
Dune::MultipleCodimMultipleGeomTypeMapper > > >,
Dune::MultipleCodimMultipleGeomTypeMapper > > > > > >, false>;
Dumux::FicksLawImplementation::ElementVolumeVariables =
Dumux::CCTpfaElementVolumeVariables,
Dumux::OnePNCVolumeVariables::NCTraits, Dumux::FluidSystems::OnePAdapter,
true>, Dumux::FluidSystems::H2OAirDefaultPolicy<>, false>, 0>,
Dumux::CompositionalFluidState,
true>, Dumux::FluidSystems::H2OAirDefaultPolicy<>, false>, 0> >,
Dumux::SolidSystems::OneCSolid, true>, Dumux::InertSolidState, true> >, double, Dumux::OnePNCModelTraits<2, true, 3> >,
Dumux::FicksLawImplementation,
Dumux::DiffusivityConstantTortuosity > > >, false>, false>;
Dumux::FicksLawImplementation::SubControlVolumeFace =
Dumux::CCTpfaSubControlVolumeFace > >,
Dumux::CCTpfaDefaultScvfGeometryTraits > > > >]’:
/home/gruenich/dune/complete/dumux/dumux/flux/cctpfa/fickslaw.hh:106:80:
  required from ‘void Dumux::FicksLawImplementation::TpfaFicksLawCache::updateDiffusion(const Problem&,
const Element&, const FVElementGeometry&, const ElementVolumeVariables&,
const SubControlVolumeFace&, unsigned int, unsigned int) [with TypeTag =
Dumux::Properties::TTag::DarcyOnePTwoC;
Dumux::ReferenceSystemFormulation referenceSystem =
Dumux::ReferenceSystemFormulation::massAveraged;
Dumux::FicksLawImplementation::Problem =
Dumux::DarcySubProblem;
Dumux::FicksLawImplementation::Element =
Dune::Entity<0, 2, const Dune::YaspGrid<2>, Dune::YaspEntity>;
Dumux::FicksLawImplementation::FVElementGeometry
=
Dumux::CCTpfaFVElementGeometry > >, false,
Dumux::CCTpfaDefaultGridGeometryTraits > >,
Dumux::DefaultMapperTraits > >,
Dune::MultipleCodimMultipleGeomTypeMapper > > >,
Dune::MultipleCodimMultipleGeomTypeMapper > > > > > >, false>;
Dumux::FicksLawImplementation::ElementVolumeVariables =
Dumux::CCTpfaElementVolumeVariables,
Dumux::OnePNCVolumeVariables::NCTraits, Dumux::FluidSystems::OnePAdapter,
true>, Dumux::FluidSystems::H2OAirDefaultPolicy<>, false>, 0>,
Dumux::CompositionalFluidState,
true>, Dumux::FluidSystems::H2OAirDefaultPolicy<>, false>, 0> >,
Dumux::SolidSystems::OneCSolid, true>, Dumux::InertSolidState, true> >, double, Dumux::OnePNCModelTraits<2, true, 3> >,
Dumux::FicksLawImplementation,
Dumux::DiffusivityConstantTortuosity > > >, false>, false>;

Re: [DuMuX] CO2 storage simulation : range of temperature in co2values.inc

[Christoph Grüninger]

Salut Etienne,

use the file
  dumux/material/components/co2tables.inc
instead. This provides data up to 400°K. The files from the examples are
smaller to reduce the runtime of the tests.

You cannot run extractproperties yourself, as it is not open source.
Someone run the software and DuMuX includes only its output.

Bye
Christoph


Am 25.04.20 um 11:16 schrieb Etienne Ahusborde:
> Dear DuMuX users,
> 
> I would like to perform a simulation for CO2 geological storage with a
> temperature equal to 358K.
> 
> Unfortunately, in the file "co2values.inc", the temperature range is
> only between 290K and 340K.
> 
> So I would like to know if it is possible for the users to generate an
> other file.
> 
> It is specified that the file was generated using ./extractproperties
> 290 340 50 10 1 495 but I can not find this function.
> 
> Thanks in advance for your recommendations.
> 
> Best regards
> 
> Etienne
> 
> ___
> Dumux mailing list
> Dumux@listserv.uni-stuttgart.de
> https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

-- 
Mit Kurzschlafwettkampfbestzeit
[Tua von Die Orsons - Sowas von egal]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] build and install Gstat

[Christoph Grüninger]

Hello Kenza,
have you deleted the build directory (build-cmake below dumux) after
installing Gstat? CMake has a heavy caching and it doesn't automatically
re-check dependencies.
So please delete dumux/build-cmake and rerun dunecontrol for dumux.

Bye
Christoph


Am 16.04.20 um 20:59 schrieb kenza bouznari:
> Hello,
> 
> I have installed the new version of dumux. Then I used the script
> '/installexternal.sh/' in /dumux///bin/ directory to install gstat.
> But it doesn't work. I guess I skip some steps in the building process !
> When I run test 1p using gstat (/make test_1p_gstat/ and then
> /./test_1p_gstat/), I get the following error : /
> /
> /This test was skipped because it failed the following CMake Conditions:
>   HAVE_GSTAT/
> Can anyone help me to fixe this issue.
> P.S. Please find attached the cmake.opts file I'm using to run cmake.
> 
> Best regards,
> 
> Kenza
> 
> ___
> Dumux mailing list
> Dumux@listserv.uni-stuttgart.de
> https://listserv.uni-stuttgart.de/mailman/listinfo/dumux
> 

-- 
Mit Kurzschlafwettkampfbestzeit
[Tua von Die Orsons - Sowas von egal]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Build broken with an error about OwnerOverlapCopyCommunication

[Christoph Grüninger]

Hi Dmitry,
you might want to try to delete all build-cmake subfolders ("rm -rf
du*/build-cmake" works for me, but be careful) and rerun dunecontrol.

This way a changed compiler will be detected and correctly configured.

Bye
Christoph


Am 30.03.20 um 22:48 schrieb Dmitry Pavlov:
> Hello,
> 
> My program's build broke all of a sudden. The error message, which I can
> not conceive, is below. My program does a 2p problem with CCFV with
> UGGrid<2>.
> 
> I know that sounds stupid, but I did not do anything. I never saw this
> error message before. The build has been OK for quite a while, but then
> I re-created the cmake-build folder to build with cmake
> -DCMAKE_BUILD_TYPE=Debug, and instantly got the error. The debug option
> probably is not the culprit because now it does not build either way.
> 
> I installed some other software recently and it might cause an update of
> some system-wide gcc packages or something. That is all I know. I will
> appreciate any pointers.
> 
> Best regards,
> 
> Dmitry
> 
> 
> 
> /usr/include/c++/7/ext/new_allocator.h: In instantiation of ‘void
> __gnu_cxx::new_allocator<_Tp>::construct(_Up*, _Args&& ...) [with _Up =
> Dune::OwnerOverlapCopyCommunication, int>;
> _Args = {const Dune::CollectiveCommunication >&,
> Dune::SolverCategory::Category&}; _Tp =
> Dune::OwnerOverlapCopyCommunication, int>]’:
> /usr/include/c++/7/bits/alloc_traits.h:475:4:   required from ‘static
> void std::allocator_traits
>>::construct(std::allocator_traits
>>::allocator_type&, _Up*, _Args&& ...) [with _Up =
> Dune::OwnerOverlapCopyCommunication, int>;
> _Args = {const Dune::CollectiveCommunication >&,
> Dune::SolverCategory::Category&}; _Tp =
> Dune::OwnerOverlapCopyCommunication, int>;
> std::allocator_traits >::allocator_type =
> std::allocator,
> int> >]’
> /usr/include/c++/7/bits/shared_ptr_base.h:526:39: required from
> ‘std::_Sp_counted_ptr_inplace<_Tp, _Alloc,
> _Lp>::_Sp_counted_ptr_inplace(_Alloc, _Args&& ...) [with _Args = {const
> Dune::CollectiveCommunication >&,
> Dune::SolverCategory::Category&}; _Tp =
> Dune::OwnerOverlapCopyCommunication, int>;
> _Alloc =
> std::allocator,
> int> >; __gnu_cxx::_Lock_policy _Lp = (__gnu_cxx::_Lock_policy)2]’
> /usr/include/c++/7/bits/shared_ptr_base.h:637:4: required from
> ‘std::__shared_count<_Lp>::__shared_count(std::_Sp_make_shared_tag,
> _Tp*, const _Alloc&, _Args&& ...) [with _Tp =
> Dune::OwnerOverlapCopyCommunication, int>;
> _Alloc =
> std::allocator,
> int> >; _Args = {const Dune::CollectiveCommunication >&,
> Dune::SolverCategory::Category&}; __gnu_cxx::_Lock_policy _Lp =
> (__gnu_cxx::_Lock_policy)2]’
> /usr/include/c++/7/bits/shared_ptr_base.h:1295:35: required from
> ‘std::__shared_ptr<_Tp, _Lp>::__shared_ptr(std::_Sp_make_shared_tag,
> const _Alloc&, _Args&& ...) [with _Alloc =
> std::allocator,
> int> >; _Args = {const Dune::CollectiveCommunication >&,
> Dune::SolverCategory::Category&}; _Tp =
> Dune::OwnerOverlapCopyCommunication, int>;
> __gnu_cxx::_Lock_policy _Lp = (__gnu_cxx::_Lock_policy)2]’
> /usr/include/c++/7/bits/shared_ptr.h:344:64:   required from
> ‘std::shared_ptr<_Tp>::shared_ptr(std::_Sp_make_shared_tag, const
> _Alloc&, _Args&& ...) [with _Alloc =
> std::allocator,
> int> >; _Args = {const Dune::CollectiveCommunication >&,
> Dune::SolverCategory::Category&}; _Tp =
> Dune::OwnerOverlapCopyCommunication, int>]’
> /usr/include/c++/7/bits/shared_ptr.h:690:14:   [ skipping 5
> instantiation contexts, use -ftemplate-backtrace-limit=0 to disable ]
> /home/dpavlov/DUMUX/dumux/dumux/nonlinear/newtonsolver.hh:1086:19:
> required from ‘std::enable_if_t<(! Dumux::isMultiTypeBlockVector()),
> bool> Dumux::NewtonSolver Comm>::solveLinearSystemImpl_(LinearSolver&,
> Dumux::NewtonSolver Comm>::JacobianMatrix&, Dumux::NewtonSolver Reassembler, Comm>::SolutionVector&, Dumux::NewtonSolver LinearSolver, Reassembler, Comm>::SolutionVector&) [with V =
> Dune::BlockVector,
> std::allocator > >; Assembler =
> Dumux::FVAssembler (Dumux::DiffMethod)0>; LinearSolver =
> Dumux::ParallelAMGBackend Dune::UGGrid<2> > >,
> Dumux::AmgTraitsImpl,
> std::allocator > >,
> Dune::BlockVector,
> std::allocator > >,
> Dumux::CCTpfaFVGridGeometry Dune::UGGrid<2> > >, false,
> Dumux::CCTpfaDefaultGridGeometryTraits Dune::UGGrid<2> > >,
> Dumux::DefaultMapperTraits Dune::UGGrid<2> > >,
> Dune::MultipleCodimMultipleGeomTypeMapper Dune::UGGrid<2> > >, Dune::Impl::MCMGFailLayout>,
> Dune::MultipleCodimMultipleGeomTypeMapper Dune::UGGrid<2> > >, Dune::Impl::MCMGFailLayout> > > >,
> (Dumux::DiscretizationMethod)2> >; Reassembler =
> Dumux::PartialReassembler (Dumux::DiffMethod)0> >; Comm =
> Dune::CollectiveCommunication; std::enable_if_t<(!
> Dumux::isMultiTypeBlockVector()), bool> = bool;
> Dumux::NewtonSolver Comm>::JacobianMatrix = Dune::BCRSMatrix 2>, std::allocator > >;
> Dumux::NewtonSolver Comm>::SolutionVector = Dune::BlockVector,
> std::allocator > >]’
> /home/dpavlov/DUMUX/dumux/dumux/nonlinear/newtonsolver.hh:1050:38:
> required from ‘bool 

Re: [DuMuX] C++17 conforming compiler required

[Christoph Grüninger]

Great news!
Having the ability to use C++17 will lead, at least in some places, to
better readable code.
Your announcement prompted me to ask the Dune developers to do the same.
And we agreed to follow DuMuX with Dune 2.8.
Cf. https://dune-project.org/community/meetings/2020-02-devmeeting/

Bye
Christoph


Am 07.02.20 um 12:50 schrieb Flemisch, Bernd:
> Dear Dumux users,
> 
> 
> Thank you for your feedback concerning the mail "Can we require C++17?"
> from the end of last year.
> 
> 
> With commit 47163f51 on master and from release 3.2 on, we require a
> C++17 conforming compiler, meaning gcc >= 7 or clang >= 5.
> 
> 
> Kind regards
> 
> Bernd

-- 
Mit Kurzschlafwettkampfbestzeit
[Tua von Die Orsons - Sowas von egal]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Precompiled headers in Dumux?

[Christoph Grüninger]

Hi Georg,
there is no Dune magic involved preventing you from using precomiled
headers. But CMake does not offer a built-in way to use them, see for
example [1] for ways to achieve what you want.

[1]
https://stackoverflow.com/questions/148570/using-pre-compiled-headers-with-cmake

Beside using PCH, you can also adjust some switches for exprt.hpp to
reduce the compilation time and pay with a slower runtime.
I assume you already have an SSD and you are using the latest Clang.

Bye
Christoph

-- 
Nous vivons une époque où les pizzas arrivent plus vite que la police.
[Claude Chabrol]

Am 03.04.19 um 09:38 schrieb georg.fut...@dlr.de:
> Hello Dumux,
> 
>  
> 
> Is there a way to include precompiled headers into Dumux? I would like
> to include the exprtk parser into my code which allows to use fancy
> mathematical expressions at runtime. However, if I do so, the
> compilation will take forever since exprtk.hpp is huge (1.4 MB).
> Therefore, the best would be to include it precompiled. I am not sure
> how to do this with Dumux/CMake/GCC. Has anyone done this before? Any
> ideas or tips would be highly appreciated.
> 
> Best regards
> Georg Futter
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Linker options to be used with DuMuX

[Christoph Grüninger]

Hi Nikolai,
that's almost pure CMake: You have to set CMAKE_EXE_LINKER_FLAGS with
your flags. Add "-DCMAKE_EXE_LINKER_FLAGS=-lvtkCommon -lvtkFiltering
-lvtkIO" to your CMAKE_FLAGS in the opts file.
Maybe you have to drop the "-l".

Then rerun dunecontrol.

Bye
Christoph

Am 04.12.18 um 18:44 schrieb Nikolai Andrianov:
> Dear DuMuX experts,
> 
> 
> Please advise how can I pass a string of linker options to DuMuX's cmake
> scripts?
> 
> 
> I would like to read the VTU output files using VTK methods, which are
> activated by passing the string ''-lvtkCommon -lvtkFiltering -lvtkIO' to
> the linker.
> 
> 
> Thank you,
> 
> Nikolai
> 
> 
> ___
> Dumux mailing list
> Dumux@listserv.uni-stuttgart.de
> https://listserv.uni-stuttgart.de/mailman/listinfo/dumux
> 

-- 
Unfortunately, plots are notoriously hard to get right. Partly, the
default settings of programs like gnuplot or Excel are to blame for
this since these programs make it very convenient to create bad plots.
-- Till Tantau, "The TikZ and PGF Packages"
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] CUDA in DUMUX

[Christoph Grüninger]

Hi Lorenzo,
you could always use another sparse solver like cuSPARSE. But you have
to check whether it is really faster. Sparse solving profits far less
from accelerator cards compared to full matrices.
In DuMuX much time is spent assembling the matrix. Out of the box
accelerators do not speed this part up. It would be great to calculate
the material laws in parallel for dozens of cells at once, but I assume
it would require a major effort. Nobody has done this far, even outside
DuMuX in the broader Dune community.

Bye
Christoph


Am 31.10.18 um 12:57 schrieb lc:
> Hello,
> 
> I'd like to ask if CUDA or any graphical processing is supported in DUMUX.
> 
> Kind regards,
> 
> Lorenzo
> 
> ___
> Dumux mailing list
> Dumux@listserv.uni-stuttgart.de
> https://listserv.uni-stuttgart.de/mailman/listinfo/dumux
> 

-- 
Unfortunately, plots are notoriously hard to get right. Partly, the
default settings of programs like gnuplot or Excel are to blame for
this since these programs make it very convenient to create bad plots.
-- Till Tantau, "The TikZ and PGF Packages"



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] memory corruption in brookscorey.hh?

[Christoph Grüninger]

Hi Edscott,
you are right, you have to recompile the complete Dune stack with Clang
and the flags for the desired saninitzer. When you use a different build
directory, it is easy to switch back to your GCC-built stack.

Hope this helps,
Christoph

Am 22.08.2018 um 18:23 schrieb Edscott Wilson:
> Hi, Christoph,
> 
> clang sounds good. I'll give it a try, But will I have to recompile dune
> with clang or will it be compatible with gcc generated .a libraries?
> 
> Best Regards,
> 
> Edscott

-- 
Unfortunately, plots are notoriously hard to get right. Partly, the
default settings of programs like gnuplot or Excel are to blame for
this since these programs make it very convenient to create bad plots.
-- Till Tantau, "The TikZ and PGF Packages"
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] memory corruption in brookscorey.hh?

[Christoph Grüninger]

Hi Edscott,
give AddressSanitizer a try, it is a great tool to find such thing you
describe. Further MemorySanitizer and Undefined Behavior Sanitizer might
turn out to be helpful, too. My best experience with these tools were
when I used the latest Clang compiler. Some of them work with recent
versions of GCC very well, too.
Valgrind might be worth a try. But it has more false positives and the
output is more difficult to understand.

Kind regards,
Christoph

Am 21.08.2018 um 18:10 schrieb Edscott Wilson:
> OK.
> I'll dig into the matter a bit further to see if I can solve where the
> problem arises. It might be an incorrect cast somewhere that screws up
> memory locations.
> 
> Best regards,
> 
> Edscott
> 
> 
> 2018-08-17 15:31 GMT-05:00 Flemisch, Bernd
>  >:
> 
> Hi Edscott,
> 
> can you please open an issue at
> https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/issues
>  ?
> Due to holiday season, it might take us some time to look at this.
> By opening an issue, it won't be forgotten.
> 
> Kind regards
> Bernd
> 
> Von: Edscott Wilson
> Gesendet: Donnerstag, 16. August, 23:47
> Betreff: [DuMuX] memory corruption in brookscorey.hh?
> An: DuMuX User Forum
> 
> 
> This is weird and should not be happening. I explain.
>  
> While debugging a generalized Dirichlet type problem, I am
> encountering a problem with the BrooksCorey material law
> (dumux/material/fluidmatrixinteractions/2p/brookscorey.hh). 
>  
> Here is the code from brookscorey.hh:
>  
> 181 using std::min;
> 182 using std::max;
> 183
> 184 swe = min(max(swe, 0.0), 1.0); // the equation below
> is only defined for 0.0 <= sw <= 1.0
> 185
> 186 return pow(swe, 2.0/params.lambda() + 3);
> 187 }
>  
> Inserting a break before the min(max()) call, I examine the value of
> swe:
>  
> Thread 1 "lswf-chem11" hit Breakpoint 2, Dumux::BrooksCorey Dumux::RegularizedBrooksCoreyParams >::krw (params=...,
>     swe=6.9319619419652626e-17) at
> 
> /opt/dune/include/dumux/material/fluidmatrixinteractions/2p/brookscorey.hh:184
> 184 swe = min(max(swe, 0.0), 1.0); // the equation below
> is only defined for 0.0 <= sw <= 1.0
> (gdb) print swe
> $11 = 6.9319619419652626e-17
>  
>  
> Then I step over the min(max() call and re-examine swe and get a
> corrupted value:
>  
> (gdb) next
> 186 return pow(swe, 2.0/params.lambda() + 3);
> (gdb) print swe
> $12 = -3.9159195083267944e+240
>  
> Stepping into the min(max()) function I see that the value which
> should be "1.0" arrives corrupted:
>  
> (gdb)
> std::min (__a=@0x7fffae00: 6.9319619419652626e-17,
> __b=@0x7fffae10: -3.9159195083267944e+240)
>     at /usr/include/c++/6.3.1/bits/stl_algobase.h:200
> 200   if (__b < __a)
> (gdb) print __b
> $16 = (const double &) @0x7fffae10: -3.9159195083267944e+240
>  
>  
> Looks like the "1.0" is being placed on the stack and being
> optimized away after the max() part completes.
>  
> Doing some changes to the code and doing the simple eff2abs law
> conversion within the regularizedbrooksCorey class template, the
> problem disappears, as the following gdb output shows:
>  
> Thread 1 "lswf-chem11" hit Breakpoint 3, Dumux::BrooksCoreyV Dumux::RegularizedBrooksCoreyVParams >::krw (params=...,
>     swe=6.9319619419652626e-17, iS=4.22627533) at
> /home/edscott/GIT/LSWF/include/2pncs/materiallaw/brookscoreyV.hh:91
> 91  swe = min(max(swe, 0.0), 1.0); // the equation below
> is only defined for 0.0 <= sw <= 1.0
> (gdb) step
> std::max (__a=@0x7fffade0: 6.9319619419652626e-17,
> __b=@0x7fffadf8: 0) at
> /usr/include/c++/6.3.1/bits/stl_algobase.h:224
> 224   if (__a < __b)
> (gdb) next
> 226   return __a;
> (gdb)
> 227 }
> (gdb)
> Dumux::BrooksCoreyV Dumux::RegularizedBrooksCoreyVParams >::krw (params=...,
> swe=6.9319619419652626e-17, iS=4.22627533)
>     at
> /home/edscott/GIT/LSWF/include/2pncs/materiallaw/brookscoreyV.hh:92
> 92  return pow(swe, 2.0/params.lambda(iS) + 3);
> (gdb) print swe
> $18 = 6.9319619419652626e-17
>  
>  
> Opinions?
>  
> Could there be something amiss in the EffToAbsLaw class template?
>  
> Or could it be a gcc bug? (using "gcc (GCC) 6.3.1 20170109" and "GNU
> gdb (GDB) 7.12.1").
>  
> I tried to use gdb within a docker container with "gcc (GCC) 7.3.1
> 20180312" and "GNU gdb (GDB) 8.1" but I get:
>  
> (gdb) run
> Starting program:
> 
> 

Re: [DuMuX] I am a new user

[Christoph Grüninger]

Hi Peleg,

just a word of advice: DuMuX (and Dune) usually require that its users
modify C++, decipher compiler messages, and lose time on programming.
Without, you won't be happy with DuMuX; better look for another code.

I know of a few cases where it worked out that a project was done
without coding in DuMuX but then others prepared the code, e.g., for a
bachelor thesis conducting a parameter study.

Bye
Christoph


Am 16.04.2018 um 15:31 schrieb Haruzi, Peleg:
> Hi Bernd,
> 
>  
> 
> Thanks for your help.
> 
> I did not get a message through the list.
> 
>  
> 
> What is shown at the third picture in the gallery is what I need, but in
> 2D, where I need it in 3D.
> 
> I am not experienced in C++. Expanding to 3D may challenging for me.
> Therefore I probably need an easier code, to not lose time on
> programming and errors coming with it.
> 
>  
> 
> I will still glad to be on the list. I may try Dumux for different type
> of simulations.
> 
>  
> 
> Kind regards,
> 
> Peleg
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Question (dune-geometry >= 2.6 )

[Christoph Grüninger]

Hi Taraneh,
please ask such question via one of the public mailing
lists. They might be of wider interest and I am nobodies
personal Dune/DuMuX help desk.

I assume your dune-grid is trunk (2.6-git) while your
dune-geometry is on the 2.5 release branch. You must
not mix different versions.

Bye
Christoph

taraneh janfada  schrieb:

Dear Christoph,
I am trying to install Dumux 2.10, but after using 
dunecontrol I see below error, if possible may I ask you what 
is the problem and how can I solve it, as I think there is 
not dune-geometry >= 2.6.
Best Regards 
Taraneh


taraneh@tobo:/temp/taraneh/dumux210> 
./dune-common/bin/dunecontrol --opts=./taranehoptim.opts all
- using default flags $CONFIGURE_FLAGS from 
/temp/taraneh/dumux210/taranehoptim.opts 
- 
ERROR: version mismatch.

   dune-grid requires dune-geometry >= 2.6,
   but only dune-geometry = 2.5.1-git is available.
Execution of dunecontrol terminated due to errors!



--
[..] Mathematicians are like theologians: we regard
existence as the prime attribute of what we study.
But unlike theologians, we need not always rely upon
faith alone.  [Lawrence Evans]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Error in Parallel computation on version 2.10 of DuMuX

[Christoph Grüninger]

Hi Etienne,
as I did never parallel runs with DuMuX myself, I can only guess.

My guess is, that your linear solver does not converge and you don't see
the error because some process is waiting for the result from a process
that terminated due to the failed convergence. So it will never
show the error or terminate.

> For the AMG solver, are there some additional parameters to give?

Do you use SuperLU or UMFPack or an iterative method after you
preconditioned with AMG? Markus Blatt often advises to use the direct
methods as the iterative ones are prone to breakdown. Might be worth a try…

Bye
Christoph

-- 
AddressSanitizer is not expected to produce false positives.
If you see one, look again; most likely it is a true positive!
[http://clang.llvm.org/docs/AddressSanitizer.html]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] curves of oil production

[Christoph Grüninger]

Hi Birger,
great news!

Is it possible to use an iterator which runs over all 
elements only once also in parallel?


Yes, that's definitely possible. You have to check for
every element whether is of ghost type.
With the new free functions, you can use the PartitionSet
elements and substract the ghosts from it, see
https://www.dune-project.org/doxygen/master/group__GIIteration.html#ga163b69ed0f5787171662b67ee84af991
and
https://www.dune-project.org/doxygen/master/namespaceDune_1_1Partitions.html#abd215a904d448bcc2525f579c2d445b8

By the way, this is more related to Dune than to DuMuX.
Asking at their mailing list might lead to better answers.

Bye
Christoph

--
On the one hand [computer "science"] has components
of the purest of mathematics and on the other hand
of the dirtiest of engineering. -- A. Oettinger 1967
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Dumux with UMFpack

[Christoph Grüninger]

Hi Ali,
SuiteSparse is package of several linear algebra algorithms including
UMFPack. They are packaged in different ways depending on the
SuiteSparse version and the Linux distribution.

1 ) You can add -lcholmod to the linker line, somewhere near the
-lumfpack line. You get the linker line with VERBOSE=1 as mentioned by
Bernd.

2) You can try building with static libraries. I haven't done this for a
long time, I am not sure whether it works out of the box or what special
flags are or avail.

3) Maybe it is worth a try to download the source code and built
SuiteSparce. Then the libraries are not oddly chopped by the distribution.

I assume this advice will not finally help. If you choose one way and
run into further troubles, please report back.

Bye
Christoph

-- 
The method has been developed for use on a high-speed electronic
computer and would be impractical for hand-solution purposes.
   [Harlow & Welch 1965]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] DUNE installation

[Christoph Grüninger]

Hi Ken,
my first guess would be, that you need to install BLAS (not openBlas)
and LAPACK and their -devel packages. Probably installing a fortran
compiler like gfortran is a good idea, too. Currently I believe, that it
should work without these dependencies, but I might be wrong. And it is
advised to have them installed anyway.

Your output is not complete, as CMake does not perform all checks,
because it reuses cached values. Remove the build-cmake directories from
dune-common and dune-istl, re-run CMake and post the complete output.
Maybe this will be more helpful.

Bye
Christoph

-- 
The method has been developed for use on a high-speed electronic
computer and would be impractical for hand-solution purposes.
   [Harlow & Welch 1965]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] 1p2C model

[Christoph Grüninger]

Hi Amine,
please have a look into our handbook. To be more precise:
In the tutorial, exercise 1 c) and its solution.

Bye
Christoph

--
Je n'ai rien contre les étrangers, quelques uns de mes 
meilleurs amis
sont des étrangers. Mais ces étrangers là ne sont pas de chez 
nous!

[Agecanonix, Astérix - Le cadeau de César]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Multidomain in parallel

[Christoph Grüninger]

Hi Georg,
sure, you can use MultiDomain / MultiDomaingrid in parallel. The best
documentation about parallel computing should be the grid howto. You can
download it from the website, see
http://www.dune-project.org/doc/grid-howto/grid-howto.pdf
or build it yourself from the dune-grid-howto module.

In DuMuX there should be tests that run in parallel, too. But not
coupled ones.

In DuMuX we introduced the MultiDomain stuff for coupling Stokes with
Darcy. Stokes leads to a saddle-point problem that could not be
efficiently solved by an iterative solver. All direct solvers are not
parallel. That's why we never tried to run the whole problem in
parallel. If your assembly is slow, it still might be worth a try.

Bye
Christoph

-- 
The method has been developed for use on a high-speed electronic
computer and would be impractical for hand-solution purposes.
   [Harlow & Welch 1965]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] New naming scheme of release branches

[Christoph Grüninger]

Dear DuMuX users and developers,
the naming scheme of our Git release branches changed. As we do not much
back-porting to release branches, this is probably not critical. If you
have scripts to check out specific branches, you have to adjust them.

All releases branches are now called
releases/X.Y
with X=2 and Y=8 for the last release.

The rationale behind this change is to become more similar to Dune and
reduce confusion in the long run.
Additionally I marked release branches as protected. They cannot be
deleted accidentally and pushing to release branches is only allowed for
the most recent one.

Bye
Christoph

-- 
The method has been developed for use on a high-speed electronic
computer and would be impractical for hand-solution purposes.
   [Harlow & Welch 1965]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Setting Properties

[Christoph Grüninger]

Hi Georg,
not sure whether there is a nice DuMuX property way to achieve what you
want, but C++11 offers std::conditional.

Bye
Christoph

-- 
The method has been developed for use on a high-speed electronic
computer and would be impractical for hand-solution purposes.
   [Harlow & Welch 1965]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] CMake support for PMOnly

[Christoph Grüninger]

Hi Vishal,
thanks for your clarification. This is none of your business, I just
added the CMakeLists.txt files.

I hope everything is going well on what you are currently doing. My
congratutions,
Christoph

-- 
The method has been developed for use on a high-speed electronic
computer and would be impractical for hand-solution purposes.
   [Harlow & Welch 1965]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Multidomain problem in 3D

[Christoph Grüninger]

Hi Georg,
the dimension should be taken from the grid.

If you want to use the Stokes discretized with the box method, it is not
guaranteed to converge, as as it uses assumptions which only holds for 2d.

Bye
Christoph

-- 
A mathematician is a scientist who can figure out anything, except
such simple things as squaring the circle and trisecting an angle.
[Evan Esar (1899-1995), Esar's Comic Dictionary]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] Whitespace cleanup on dumux-devel this weekend

[Christoph Grüninger]

Dear DuMuX users,
this email is only relevant, if you are using dumux-devel.

This weekend we will enable the whitespace-hook for our 
dumux-devel Git
repository. This will prevent us from committing sources with 
tabs or
trailing whitespace. As a host of source file does not comply 
to this
rule, many commits will be rejected. To keep the disruption 
minimal,
we will clean up all source file from dumux-devel master this 
weekend

before enabling the hook.

To reduce you merging efforts, you might consider committing 
all local

changes until Friday midnight (November 5th, 23:59).

You can get more details from FS#272
http://www.dumux.org/flyspray/index.php?do=details_id=272

Bye
Christoph

--
Je n'ai rien contre les étrangers, quelques uns de mes 
meilleurs amis
sont des étrangers. Mais ces étrangers là ne sont pas de chez 
nous!

[Agecanonix, Astérix - Le cadeau de César]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Triangular prism mesh

[Christoph Grüninger]

Hi Jelena,
the crucial point is the grid with prism elements. Half a year ago,
Martin tried to handle the elements with dune-prismgrid [1] module.

Unfortunately the RC1 tarball from the website works only with Dune 2.2
which is ancient. We got access to the Git repository from Christoph
Gersbacher, and with some patches from our side we got it running with
Dune 2.3. With 2.4 we don't know how much work is to be done. You can
ask Christoph Gersbacher to get access to the repository and we can
share our patches. But we don't support prismgrid.

Best wishes,
Martin Beck & Christoph

[1] http://dune.mathematik.uni-freiburg.de/grids/dune-prismgrid/

-- 
A mathematician is a scientist who can figure out anything, except
such simple things as squaring the circle and trisecting an angle.
[Evan Esar (1899-1995), Esar's Comic Dictionary]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Triangular prism mesh

[Christoph Grüninger]

Hi Jelena,
that depends on what you want to do. A cell-centered finite-volume
scheme should work. Out of my head I don't know how to create a grid
with triangular prisms, but people did that already with some Dune grids
for sure.

But there is no example and will require some work from your side. And
probably you cannot use the box scheme on prisms.

Bye
Christoph

-- 
A mathematician is a scientist who can figure out anything, except
such simple things as squaring the circle and trisecting an angle.
[Evan Esar (1899-1995), Esar's Comic Dictionary]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] UMFPack

[Christoph Grüninger]

Hi Georg,
not sure about your problem. Have you reconfigured and re-build dune-istl?
In any case, try the ISTL tests using UMFPack. If they fail, it is not a
DuMuX issue. What about the DuMuX test cases using UMFPack?

Bye
Christoph

-- 
A mathematician is a scientist who can figure out anything, except
such simple things as squaring the circle and trisecting an angle.
[Evan Esar (1899-1995), Esar's Comic Dictionary]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] Removal of Autotools

[Christoph Grüninger]

Dear DuMuX users,
I just removed the support for Autotools. Please use CMake, especially
when you are using Dune master (will be 2.4). If you hesitate to use
CMake, stay with revision 14916.

I will clean up dumux-lecture in the near future, dumux-devel will
remain untouched for the time being.

Bye
Christoph

--
Schauen Sie sich Stuttgart an: ein Drecksloch, ein städtebaulicher
Irrtum, ein zubetonierter Talkessel, der von den Abgasen einer ewig im
Stau stehenden Blechlawine aufgeheizt wird. [ARD-Tatort vom 21.06.2015]



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Check for phase presence in the mpnc model

[Christoph Grüninger]

Hi Georg,
I switched to UMFPack, it is better* and also a direct solver. Pardiso
could be used, but you need a license.
Have a look at at one of our Stokes examples how to use UMFPack.

* Faster, terminates more reliably with singular matrices and
subjectively converges better.

Bye
Christoph

-- 
People who are more than casually interested in computers should have
at least some idea of what the underlying hardware is like. Otherwise
the programs they write will be pretty weird. -- Donald Knuth



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] output the pressure, the saturation at specific location of the domain

[Christoph Grüninger]

Dear Hai,
have a look at the addOutputVtkFields method in the models. 
There you
have access to the time manager and the current solution. 
Iterating

over the grid might cost you a bit of performance.

Bye
Christoph

--
Je n'ai rien contre les étrangers, quelques uns de mes 
meilleurs amis
sont des étrangers. Mais ces étrangers là ne sont pas de chez 
nous!

[Agecanonix, Astérix - Le cadeau de César]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] strange artifacts

[Christoph Grüninger]

Hi Georg,
can you provide the code where you set the boundary condition type?

Bye
Christoph

-- 
People who are more than casually interested in computers should have
at least some idea of what the underlying hardware is like. Otherwise
the programs they write will be pretty weird. -- Donald Knuth



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] Minimal compiler versions for DuMuX 2.8

[Christoph Grüninger]

Dear DuMuX users,
we want to tighten the supported compiler versions for the next DuMuX
release 2.8. We aim for GCC 4.7+ and Clang 3.3+.
Is there any reason to changes the supported compiler version?

Thanks for sharing your thoughts.

Bye
Christoph

-- 
People who are more than casually interested in computers should have
at least some idea of what the underlying hardware is like. Otherwise
the programs they write will be pretty weird. -- Donald Knuth



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] How to write only every n(th) time step to output file in 2p2cni implicitproblem

[Christoph Grüninger]

Hi Jahnes,
please do not write emails to dumux-owners, only use the mailing list.

There is a parameter, that can be set to write only every n-th time
step, but I cannot find it in the parameter list
http://www.dumux.org/doxygen-stable/html-2.6/a01137.php

What's wrong with setoutputinterval(nth)?

If you really want define time interval, for example very 100th year,
you need to set episodes of that length. Actually, I don't know exactly
how that is done.

Bye
Christoph


Am 03.04.2015 um 21:31 schrieb Giannis Varvaris:
 Dear Dumux users,
 
 I am trying to get familiar with Dumux and I want to ask you if it is
 possible to help me for solving a problem I have. In 2p2cni, I try to
 simulate my problem for 10.000 years and for that reason I want to write
 the results to output file not for every single step but depending on
 the time (for the first 10 years every time step, for the next 100 years
 every 20 time steps etc). I read the Dumux Handbook and for the
 ex2tutorialproblem you have used the command setoutputinterval(nth). I
 would like to ask you which is the appropriate command for that case for
 the 2p2cni problem.
 
 Thank you in advance
 
 Jahnes

-- 
People who are more than casually interested in computers should have
at least some idea of what the underlying hardware is like. Otherwise
the programs they write will be pretty weird. -- Donald Knuth



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] DuMuX user meeting program outline

[Christoph Grüninger]

Dear participants of the upcoming DuMuX user meeting!

We ask all participants to tell us
- your email address to keep you informed about the event
- your affiliation (University, company etc.)
- title of your presentation

The program outline:
- Every participant has a 25 minute slot with additional 5 minutes for
  questions.
- Members of LH2 will present specific DuMuX related topics or their
  current work. The time slots will be shorter like 10 to 15 minutes and
  is not yet decided.
- There will be discussions about DuMuX and Dune.
- If you have questions, that are useful to a wider audience, email
  them to us; we will address these topics.

All information can be found at
http://www.dumux.org/dumuxusermeeting2015.php

If you have questions, don't hesitate to contact us.

Best wishes,
Bernd Flemisch and Christoph Grüninger



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] Installing Dune and DuMuX

[Christoph Grüninger]

Hi JF,
to use dumux-lecture, you don't have to install it. Usually I just put
it in a directory, next to the Dune modules and DuMuX. Then run
dunecontrol and you are done.

dune-howto has been updated regularly, so you can use it. Other guides
for installation can be found at:
- DuMuX handbook
- http://www.dumux.org/installation.php
- http://www.dune-project.org/doc/installation-notes.html

Be aware that when using Dune from Git, that we recently have changed
the build system which is not yet reflected in these documents.
I advice you anyway to use Dune 2.3 and DuMuX 2.6 releases.

Have fun with Dune,
Christoph

-- 
GDB does hate your application, expresses its contempt through the
design of its command-line interface.   -- Tom Tromey, FOSDEM 2014



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] convergence rate

[Christoph Grüninger]

Hello Hai,
what do you mean by weak and strong convergence rate?

A student of my checked the L2 error for refined grids and time steps
with the Richards equation and the FV method. We got the expected first
order convergence.
You can find the detail in his Bachelor's thesis, but it is German
Baccour, M.H.: Verifikationsproblem mit bekannter Lösung für die
Zwei-Phasen-Darcy-Gleichung

Additional there was a Master's thesis by Markus Wolff, it's written in
English and might help. I haven't checked myself, Bernd told me that:
Wolff, M.: Comparison of mathematical and numerical models for twophase
flow in porous media.

Just out of curiosity: What's your affiliation?

Bye,
Christoph

-- 
GDB does hate your application, expresses its contempt through the
design of its command-line interface.   -- Tom Tromey, FOSDEM 2014



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] install target

[Christoph Grüninger]

Hi Mladen,
thank you for your hint and the patch, you are right. I committed your
changes and backported them to the 2.6 release branch.

Bye
Christoph

-- 
GDB does hate your application, expresses its contempt through the
design of its command-line interface.   -- Tom Tromey, FOSDEM 2014
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Buildsystem and compiler problems

[Christoph Grüninger]

Hi Markus,

 It is patch r13248 in
 svn://svn.iws.uni-stuttgart.de/DUMUX/dumux/branches/strip-pdelab .
 
 Together with r13246 out-of-source builds work. I still get build
 errore but theses seem to be compiler related.

I suggest to include this patch when you'll merge your branch.
Thanks again for investigation the problem.

Bye
Christoph

-- 
GDB does hate your application, expresses its contempt through the
design of its command-line interface.   -- Tom Tromey, FOSDEM 2014



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Buildsystem and compiler problems

[Christoph Grüninger]

Hi Markus,
aah out of source builds. I was wondering what you do so special to
trigger this problem. As we have Ubuntu users which should test the
Debian specific stuff, I ran out of ideas.

Your patch will make the Makefile.am's even less readable, but as most
DuMuX users copymodify these files, I don't care. The downside you
mention is negligible.
If nobody objects, I'd be happy to review your patch and welcome your
commit.

Bye
Christoph

-- 
GDB does hate your application, expresses its contempt through the
design of its command-line interface.   -- Tom Tromey, FOSDEM 2014



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Fwd: no dunecontrol under usr/local/lib/

[Christoph Grüninger]

Hi,
quoting the last two lines:

 --- Failed to build dune-multidomaingrid ---
 Terminating dunecontrol due to previous errors!

Why do you need multidomaingrid? If you have no answer to that, remove
it and try again.
If you need it, try to fix it, it seems to be broken. Usually the
missing file configure.ac is provided. So, difficult to tell what
exactly went wrong with your multidomaingrid.

Bye
Christoph


-- 
GDB does hate your application, expresses its contempt through the
design of its command-line interface.   -- Tom Tromey, FOSDEM 2014



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Fwd: no dunecontrol under usr/local/lib/

[Christoph Grüninger]

Hi 吴昊,
last time, multidomaingrid was definitively the problem. This time
dumux.lecture has the same problem. I don't know how you get the module,
but it is outdated.

 lecture/mm/Makefile.am:2: warning: whitespace following trailing backslash

I fixed that back in 2013 (revision 11944, 2013-11-10).

So please checkout the latest release version/tagged version.

 ERROR: configure.[in|ac] found, but configure missing.

For sure, this file is part of lecture since we forked dumux-lecture
from dumux. Either the package you got is broken or something went wrong
during Subversion checkout.

How did you get dumux-lecture?

Bye
Christoph

-- 
GDB does hate your application, expresses its contempt through the
design of its command-line interface.   -- Tom Tromey, FOSDEM 2014



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Fwd: no dunecontrol under usr/local/lib/

[Christoph Grüninger]

Hi 吴昊,
the PaxHeader stuff can be related to encodings and 
decompression
with tar, see
http://www.gnu.org/software/tar/manual/html_section/Portability.html

Try to decompress the dune modules without encoding struggles.

Bye
Christoph

-- 
Das Kommt darauf an, in welchen Zeiträumen du
rechnest. Für Geologen: ja, in naher Zukunft
[Stefan Seyfried in suse-laptop]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Using ALUGrid in MpNc

[Christoph Grüninger]

Hi Mojdeh,
I tried to reproduce your problem, but I could not apply
your patches, neither to DuMuX trunk nor to release 2.5.
What version do you use? Would you mind to use Dune 2.3
and DuMux 2.5?

Your changes looks ok, but I am wondering why you include
the hassle with the inlet / boundary ids. Why do you want
to use ALUGrid? Maybe there is an other example with
ALUGrid more suitable?

Concering Markus' question
 Maybe he is just missing the ALUGRID compile/linker flags?

I don't think so, as we never the flags explicitly, but
all executable use the PKG_ALL flags. I think that is done
somewhere in the configure.ac.

Bye
Christoph

-- 
Das Kommt darauf an, in welchen Zeiträumen du
rechnest. Für Geologen: ja, in naher Zukunft
[Stefan Seyfried in suse-laptop]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Buildsystem and compiler problems

[Christoph Grüninger]

Hi Markus!
 thanks for the patch.

Thanks for your feedback. Was this the only problem with
GCC 4.6? If yes, then I'll apply my patch.

 I guess it works with GNU make (gmake) and you have maybe
 some other make flavor?
 
 I do have gnumake:

No good ideas left. Can you send my your Makefile by private
mail? Maybe a direct comparison with my local file will 
provide
an insight.

Bye
Christoph

-- 
Das Kommt darauf an, in welchen Zeiträumen du rechnest. Für 
Geologen:
ja, in naher Zukunft :-) [Stefan Seyfried in suse-laptop]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Using ALUGrid in MpNc

[Christoph Grüninger]

Hi Mojdeh,
all looks fine, at least no obvious mistake. Maybe the change
from a one-dimensional grid to a two-dimensional went wrong.
Can you send the complete error message from your compiler or
the output from the program?

Bye
Christoph

-- 
Das Kommt darauf an, in welchen Zeiträumen du
rechnest. Für Geologen: ja, in naher Zukunft
[Stefan Seyfried in suse-laptop]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Eigenvalues of the jacobian matrix

[Christoph Grüninger]

Hi Georg,
as far as I know, you can not get the eigenvalues out of the box.
Dune-ISTL has some eigenvalue computation support.
I'd advice you to hack the DuMuX assembler and print everything into a
Matlab/Octave file. Using their capabilities it's easy to obtain the
eigenvalues.
- you can get the matrix for example in
dumux/implicit/common/implicitassembler.hh at the end of the assemble()
method
- use dune-istl's Dune::writeMatrixToMatlab

Bye
Christoph

-- 
... as Sir Cyril Hinshelwood has observed ...fluid dynamicists
were divided into hydraulic engineers who observed things that
could not be explained and mathematicians who explained things
that could not be observed. -- James Lighthill
*
CMWR 2014: 10th - 13th June 2014 in Stuttgart
 Please visit www.cmwr14.de
*



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Problem with exercise 2

[Christoph Grüninger]

Hi Khaled,
please keep the emails on the mailing list. This will help future
readers and increases the number of people who can answer you.

You are right, the patch for the tutorial_coupled.cc is not provided. I
missed this fact.
I had a closer look on what you tried. I believe your problem is that
you copied the body of your main function instead of replacing it.
Please delete the 4 lines after the declaration of main and keep the
lines where you pointed with the green arrow in your PDF.

If it does still not work, please attach the compiler's error message.

Bye
Christoph


Am 28.03.2014 19:40, schrieb Khaled Maaloum:
 hi Christoph
  But the folder  dumux/tutorial/solutions_coupled/  does not present the
 changes on the folder  tutorial_coupled.cc

-- 
... as Sir Cyril Hinshelwood has observed ...fluid dynamicists
were divided into hydraulic engineers who observed things that
could not be explained and mathematicians who explained things
that could not be observed. -- James Lighthill
*
CMWR 2014: 10th - 13th June 2014 in Stuttgart
 Please visit www.cmwr14.de
*
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] Problem with exercise 2

[Christoph Grüninger]

Hi Khaled,
welcome to DuMuX. Have you had a look at the patches in
dumux/tutorial/solutions_coupled/
which offer you the solutions to the exercises? If the right patch does
not answer your question, please do not hesitate to ask again.

Best,
Christoph

-- 
... as Sir Cyril Hinshelwood has observed ...fluid dynamicists
were divided into hydraulic engineers who observed things that
could not be explained and mathematicians who explained things
that could not be observed. -- James Lighthill
*
CMWR 2014: 10th - 13th June 2014 in Stuttgart
 Please visit www.cmwr14.de
*



signature.asc
Description: OpenPGP digital signature
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] grid loadBalance()

[Christoph Grüninger]

Hi Ali,
as I never used grid partitioning or load balancing myself, I can only
guess. My tips are:
1) are you using Metis or Parmetis to create the partitions? Maybe they
can be tweaked to generate better results.
2) I think you have to create a more coarse grid, create the partitions
and refine then. Eike Mueller did that, he started with number of cells
= number of processors. Maybe you can find some hint in the Dune mailing
list archive.

Bye
Christoph

-- 
Une science n'était vraiment développée que quand elle
pouvait utiliser les mathématiques.(Paul Lafargue)
*
CMWR 2014: 10th - 13th June 2014 in Stuttgart
 Please visit www.cmwr14.de
*
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] DuMuX with Intel compiler

[Christoph Grüninger]

Hi Ali,
as the Intel compiler is not officially supported in Dune 2.2 and I
never tried one myself, it could be possible that you have to fix some
issues yourself.

From your output I would advise you to do:
- use --enable-fieldvector-size-is-method in your configure flags. This
will reduce the warnings and switch off the Dune 2.1 behavior. In Dune
2.3 this will be default.
- I am not sure about the errors from ALUGrid. The error message is not
very helpful without investigating the code more than I did.
- The catastrophic error about the missing unordered_map is related to
the not enabled C++11 support. You have to add -std=c++11 to the compile
flags. dune-common tries to detect this, but in Dune 2.2 only -std=c++0x
is tested that does not work with Intel. Either add -std=c++11 to the
CXX_FLAGS or replace the file m4/cxx0x_compiler.m4 with the one from
dune-common's master where the test is exted to -std=c++11.

When you use C++11 with ipcp, you might run into this bug:
https://dune-project.org/flyspray/index.php?do=detailstask_id=1228project=1
If this is the case, you can use the code from the bug tracker's last
comment (from Markus). I would like to get feedback, whether this solved
the bug for you.

Bye
Christoph

-- 
Une science n'était vraiment développée que quand elle
pouvait utiliser les mathématiques.(Paul Lafargue)
*
CMWR 2014: 10th - 13th June 2014 in Stuttgart
 Please visit www.cmwr14.de
*
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] SuperLU and UG version

[Christoph Grüninger]

Hi Ali,
Dune 2.2.1 supports UG up to Patch8. Newer versions of UG are only
supported by dune-master (will be 2.3) or in the releases/2.2 branch
after the 2.2.1 tag.


SuperLU 4.3 is fully supported. If it is either installed or you have
set --with-superlu=superlu-path --with-superlu-lib=libraryname
correctly, it should work. If not, you can send us the config.log to
check what goes wrong.

Bye
Christoph

-- 
Une science n'était vraiment développée que quand elle
pouvait utiliser les mathématiques.(Paul Lafargue)
*
CMWR 2014: 10th - 13th June 2014 in Stuttgart
 Please visit www.cmwr14.de
*
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] How to install DUNE and DuMuX

[Christoph Grüninger]

Hi Okpu,
Dune 2.2 does not support CMake. Anywhere else like in trunk or the 
2.2-cmake branches CMake support is still experimental. I guess you 
checked out the 2.2-cmake branches of the other Dune modules but missed 
it for dune-localfunctions.
As far as I know, the 2.2-cmake branches do not work out of the box with 
DuMuX 2.3 but you have to use DuMuX from trunk.

Try to use the autotools which is done automatically when you call 
dunecontrol. I don't see why you refuse using dunecontrol. Its use is 
described in our documentation and in one of my past answers to you.

Bye
Christoph

-- 
Digital information lasts forever – or five years, whichever comes first.
-- Jeff Rothenberg, 1997
*
CMWR 2014: 10th - 13th June 2014 in Stuttgart
  Please visit www.cmwr14.de
*
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] dofMapper not declared in implicitmodel.hh

[Christoph Grüninger]

, Dune::FieldVectordouble, 2 const, Dune::FieldVectordouble,
 2 const)in main.o
Dune::SGrid2, 2, double::makeSGrid(Dune::arrayint, 2ul
 const, Dune::FieldVectordouble, 2 const, Dune::FieldVectordouble,
 2 const)in main.o

Dune::Exception::_hook, referenced from:
Dune::Exception::Exception()in main.o
Dune::Exception::Exception()in main.o
Dune::Exception::Exception()in main.o
 ld: symbol(s) not found
 collect2: ld returned 1 exit status


 Any help please ?

 2013/5/2 Christoph Grüninger christoph.gruenin...@iws.uni-stuttgart.de
 mailto:christoph.gruenin...@iws.uni-stuttgart.de

 Hi Errazi,
 I never used Xcode myself, so I am not sure what to do. Have you
 compiled and linked th gettingstarted.cc with the Makefile from the
 autotools or is it generated by Xcode?

 Please try using the Makefile from a shell.

 The error indicates, that libdunecommon is not included in the linking.

 Bye
 Christoph


-- 
The Romans didn't find algebra very challenging,
because X was always 10.

CMWR 2014: 9th - 13th June 2014 in Stuttgart
  Please visit www.cmwr14.de

___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] dofMapper not declared in implicitmodel.hh

[Christoph Grüninger]

Hi Errazi,
could you please stay on the list instead of writing the emails to me 
directly?

Xcode is not a compiler but a development environment that can make use 
of various compilers. So it is not clear enough. Unless you can name us 
your compiler, we are here at a dead end.

I propose you stop using Xcode and use GCC (or a Clang newer then 3.0) 
directly from a terminal. There it's easy to determine the compiler. If 
you get DuMuX compiling there, it is up to you to configure your Xcode 
to whatever pleases you.

Christoph


Am 03.05.2013 10:05, schrieb Errazi BEN AHMED:
 Hi Christoph,

 Actually, I'm using files from SVN as described in the Dumux
 installation page (http://www.dumux.org/installation.php) and
 followed exactly the instructions described in that page, arriving at
 the bottom the example is not working for me ( previous mail) and all
 this is before using Xcode or any other compiler.
 hope that's clear enough right now.

 regards
 Razi

-- 
The Romans didn't find algebra very challenging,
because X was always 10.

CMWR 2014: 9th - 13th June 2014 in Stuttgart
  Please visit www.cmwr14.de

___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] Problem with dune-localfunctions

[Christoph Grüninger]

Hi Errazi,
you are right, that shouldn't be a problem. I just checked with 
dune-localfunctions
2.2.1 tarball and it works fine for me. Try to re-download the tarball (or check
it out from SVN) and run dunecontrol again.

If nothing helps, you can exclude the folder from being configured.
- remove from dune-localfunctions/configure.ac the line containing
 dune/localfunctions/brezzidouglasmarini/brezzidouglasmarini1q2d/Makefile
- change in 
dune-locaalfunctions/dune/localfunctions/brezzidouglasmarini/Makefile.am from
  SUBDIRS = brezzidouglasmarini12d brezzidouglasmarini1q2d
to
  SUBDIRS = brezzidouglasmarini12d

Then rerun dunecontrol.

Hope that helps,
Christoph

-- 
[..] we shall send to the moon, 240,000 miles away from the control
station in Houston, a giant rocket [..] and then return it safely to
earth, re-entering the atmosphere at speeds of over 25,000 miles per
hour, causing heat about half that of the temperature of the sun,
almost as hot as it is here today, [..] then we must be bold.
[John F. Kennedy Moon Speech - September 12, 1962 - Rice Stadium]
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] DuMuX Installation on Mac

[Christoph Grüninger]

Hi Errazi,
probably makegcc is a typo where you forgot a space between make and gcc?
You can also install the packages one by one, you don't need to use them 
all in a single command.

Bye
Christoph

Am 30.04.2013 08:47, schrieb Errazi BEN AHMED:
 Hello,

 thank you for the brief explanation,
 I did what you said till sudo port install libcxx and every thing was
 working fine, then coming to the installation page
 I tried this sudo port install automake libtool make gcc gcc-fortran
 gcc-c++ subversion and gor Error: Port make not found
 I asked for help from someone and he said that I should type this
 unstead: sudo port install automake libtool make gcc gfortran g++
 subversion with that it started calculating for few minutes but gave
 this error: Port makegcc not found.

 Do you have any idea how to fix that ?

 thank you

-- 
The Romans didn't find algebra very challenging,
because X was always 10.

CMWR 2014: 9th - 13th June 2014 in Stuttgart
  Please visit www.cmwr14.de

___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Dumu Installation on Mac

[Christoph Grüninger]

Hi Errazi,
your last email did not help me. Maybe you need more background to find 
the problem yourself.
dunecontrol identifies dune-modules by looking for dune.module files 
which are then parsed.

.Trash/dune-common-2.2.1/dune-common-2.2.1/PaxHeaders.32165/dune.module 
does not contain a Module entry
Execution of dunecontrol terminated due to errors! 

I assume you have deleted some modules and your hidden trash folder 
still contains the dune.module file. Maybe you move the dune modules 
somewhere else, where the .Trash is no subfolder of the dune folder. Or 
you remove all dune.module files from your trash.

You can find more information about the build system in the DuMuX 
handbook and the build system howto from the Dune homepage.

Hope that helps
Christoph

-- 
The Romans didn't find algebra very challenging,
because X was always 10.

CMWR 2014: 9th - 13th June 2014 in Stuttgart
  Please visit www.cmwr14.de

___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

[DuMuX] DuMuX requires patching to work with Dune-trunk

[Christoph Grüninger]

Dear DuMuX users,
the recent changes in Dune-trunk (dune-common r7429) require to
patch DuMuX. Otherwise modules that require DuMuX cannot find
DuMuX. If you are using Dune-trunk, patch your local copy by
executing inside the dumux folder:
  patch -p0  ../patches/dumux-m4.patch

After a Dune 2.3 release we will apply the changes to DuMuX-trunk
and will break compatibility to Dune versions older than 2.3.

Greetings
Christoph
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] DUMUX SOFEWARE and Further Help in my Research Work

[Christoph Grüninger]

Hi Okpu,
what's wrong with http://www.dumux.org/download.php ? Enter
your name, organization and your email address and you can
download a tarball. Alternatively you can get DuMuX via 
Subversion.

Have a look into our handbook
http://www.dumux.org/documents/dumux-handbook-2.2.pdf
it helps you to get started with DuMuX, dune and all 
related stuff.

If you have further questions, please ask via the mailing
list dumux@listserv.uni-stuttgart.de as mentioned on our
home page.

Bye
Christoph


ambrose OKPU onnexcel...@yahoo.co.uk schrieb:


Hi,

Goodmorning. I am a research student in Heriot Watt University, Edinburgh, 
UK. Presently, I am working on a project titled Developing a Model for 
CO2 - EOR in a porous media and its application in Niger Delta, Nigeria. 
I was looking out for sofeware that can be useful, which i discovered DUMUX 
but i have being having challenges in downloading the files.
Also, i will like to know the possible means of getting the sofeware since i 
did not see any link in download it. Alternatively, I don't mind coming to 
your university to see you in person because it will create an avenue for 
proper discussion as regards this.

On this note, I want to say thank you and looking forth to hear from you.

Regards,
Okpu Onne Ambrose.

___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux

Re: [DuMuX] Stuck with boundary condition setting

[Christoph Grüninger]

Hi Romain,
are you sure, that you use the right x-values in the method 
boundaryTypes(..)? Try using the onUpperBoundary_(..) as several other 
test cases use them.

You do not need to change anything in neumann(..), because in 
bounadryTypes(..) you specified the injectors to be of Dirichlet type. 
Add the right value for dirichlet(..) and it should work.

Salut
Christoph
___
Dumux mailing list
Dumux@listserv.uni-stuttgart.de
https://listserv.uni-stuttgart.de/mailman/listinfo/dumux