Ecosyste.ms: Timeline
Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.
mathrack created a comment on a pull request on xcompact3d/Incompact3d
I think we could also move some stretching-specific parameters to the new module : `alpha`, `beta`, ...
mathrack opened a pull request on xcompact3d/Incompact3d
Move stretching suborutines to a dedicated module and fix undefined y…
…praf values Closing #293
mathrack created a branch on xcompact3d/Incompact3d
fix_ypraf_ibm - Current CPU version of our solver for the Navier-Stokes equations
mathrack created a comment on an issue on xcompact3d/Incompact3d
Not really, catching this issue with gdd / valgrind is a bit difficult because it can change depending on the compiler and number of CPUs used... This issue was discussed in https://github.com/x...
mathrack created a comment on an issue on xcompact3d/Incompact3d
Ok, I will start a draft for a fix shortly !
mathrack created a comment on an issue on xcompact3d/Incompact3d
If the code supports periodicity + stretching, we could fix the IBM issue with a modification of the stretching subroutine. However, this looks a bit complex and the `poisson_000` subroutine seems ...
mathrack created a comment on a pull request on 2decomp-fft/2decomp-fft
Hello, If the DCT / DST in this PR is working, it could replace the current one. In this situation, conflicts are expected. I have been working on this PR for a very long time and I can not f...
mathrack pushed 1 commit to dev 2decomp-fft/2decomp-fft
- Update modules for Intel CI (#356) * Update modules for Intel CI * Add missing file from src list for exe fft_mul... 40768f6
mathrack closed a pull request on 2decomp-fft/2decomp-fft
Update modules for Intel CI
Fix the CI for Intel.
mathrack created a comment on a pull request on 2decomp-fft/2decomp-fft
Do you need a changelog since last release ?
mathrack opened a pull request on 2decomp-fft/2decomp-fft
Fix IO for reduced precision and 3D array in y / z pencil
Closing #354
mathrack created a branch on 2decomp-fft/2decomp-fft
mathrack-patch-1 - The 2DECOMP&FFT library provides access to a slabs and pencil decompositions as well as FFTs.
mathrack created a comment on an issue on 2decomp-fft/2decomp-fft
Running the following commands : ``` FC=mpif90 cmake -S ./ -B ./build -DCMAKE_BUILD_TYPE=dev -DBUILD_TESTING=on -DFFT_Choice=fftw_f03 -DFFTW_ROOT=~/opt/fftw3/fftw-3.3.10_bld/ -DCOMPLEX_TESTS=OFF...
mathrack created a comment on an issue on 2decomp-fft/2decomp-fft
If the library is compiled with double precision and the preprocessor variable `SAVE_SINGLE` is defined, the issue can be triggered. This issue is present only in the branch main. I think it is not...
mathrack opened an issue on 2decomp-fft/2decomp-fft
[main] - Issue when writing a 3D array in y/z pencil with reduced precision using mpiio_write_real_coarse
The allocate statement is valid only when the input is in x pencil. It is not valid if the input is in y or z pencil https://github.com/2decomp-fft/2decomp-fft/blob/4ff8f79098e6028dad6efd9d994b6...
mathrack created a review on a pull request on xcompact3d/Incompact3d
The PR should be limited to the (CMake) testing framework. If possible, please avoid updating / deleting input files here.
mathrack created a review on a pull request on xcompact3d/Incompact3d
The PR should be limited to the testing framework. If possible, please avoid updating / deleting input files here.
mathrack created a review comment on a pull request on xcompact3d/Incompact3d
The scope of the PR is to split the tests. I think updating the input files is outside of the scope of this PR.
mathrack created a review comment on a pull request on xcompact3d/Incompact3d
If it is too much work, keep it for later
mathrack created a review comment on a pull request on xcompact3d/Incompact3d
Having the `partype` type private and hidden from the rest of the code forces the particle development to stay inside a single module / file. On one hand, it seems easier because everything is loca...
mathrack created a review comment on a pull request on xcompact3d/Incompact3d
I think the target attribute is needed if the subroutine contains something like `xxx => pa` or `xxx => pa%element`. This is probably not needed here.
mathrack created a review on a pull request on xcompact3d/Incompact3d
I have suggested some modifications. I have not yet looked at `particle.f90`. I have an error with the following build command : ``` rm -rf ./build/ && FC=mpif90 cmake -S ./ -B ./build -DBUILD_T...