- Nov 05, 2020
-
-
Stephan Seitz authored
of stencil
-
- Oct 30, 2020
-
-
Stephan Seitz authored
-
Stephan Seitz authored
-
- Oct 28, 2020
-
-
Stephan Seitz authored
-
- Oct 01, 2020
-
-
Stephan Seitz authored
-
- Feb 24, 2020
-
-
Stephan Seitz authored
-
Stephan Seitz authored
-
- Feb 17, 2020
-
-
Stephan Seitz authored
-
- Feb 06, 2020
-
-
Stephan Seitz authored
-
- Feb 04, 2020
-
-
Stephan Seitz authored
-
- Jan 31, 2020
-
-
Stephan Seitz authored
-
- Jan 29, 2020
-
-
Stephan Seitz authored
-
Stephan Seitz authored
-
Stephan Seitz authored
-
Stephan Seitz authored
-
- Jan 28, 2020
-
-
Stephan Seitz authored
-
- Dec 06, 2019
-
-
Martin Bauer authored
error occured when using folders that can be accessed with two different paths via symlinks
-
- Dec 01, 2019
-
-
Martin Bauer authored
adapt to new create_staggered_kernel API See merge request pycodegen/pystencils_walberla!3
-
- Nov 29, 2019
-
-
Michael Kuron authored
-
- Nov 17, 2019
-
-
Martin Bauer authored
codegen.py fix: bug in the communication of corner directions See merge request pycodegen/pystencils_walberla!2
-
- Nov 10, 2019
-
-
Markus Holzer authored
In the function comm_directions the corner directions were treated wrongly. This fix also generalized comm_directions so it will also work for higher dimensions
-
- Oct 17, 2019
-
-
Martin Bauer authored
-
- Oct 06, 2019
-
-
Martin Bauer authored
-
Martin Bauer authored
-
- Jul 26, 2019
-
-
Martin Bauer authored
-
- Jul 11, 2019
-
-
Martin Bauer authored
-
Martin Bauer authored
-
- Jun 18, 2019
-
-
Martin Bauer authored
-
Martin Bauer authored
- when reading from ghost layers (previously the only option) a pull communication is required - if a kernel writes to the ghost layer a "push" communication has to be done - the new PackInfo generator can now derive push and pull packinfos from a given kernel
-
- Jun 07, 2019
-
-
Martin Bauer authored
-
- May 06, 2019
-
-
Martin Bauer authored
-
- May 05, 2019
-
-
Martin Bauer authored
-
- May 03, 2019
-
-
Martin Bauer authored
-
- Apr 28, 2019
-
-
Martin Bauer authored
-
Martin Bauer authored
-
- Apr 25, 2019
-
-
Martin Bauer authored
-
- Apr 24, 2019
-
-
Martin Bauer authored
- turned on restrict keyword by default (makes large difference on GPUs) - smarter block indexing: changing block size depending on domain size Example: previously there where (1,1,1) blocks when requested block size was (64, 1, 1) and domain size (1, 512, 512), now the block size is changed automatically to (1, 64, 1) in this case - added __lauch_bounds__ to kernels to allow better optimizations from the CUDA compiler
-
- Apr 18, 2019
-
-
Martin Bauer authored
- missing pragma once added - static variable to member in overlap sweep, when called with changing block sizes this lead to wrong results
-
- Apr 16, 2019
-
-
Martin Bauer authored
-
- Apr 04, 2019
-
-
Martin Bauer authored
-