CP2K Open Source Molecular Dynamics

User Tools

Site Tools

Trace:
dev:todo

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revisionBoth sides next revision
dev:todo [2017/02/10 09:31] – oschuettdev:todo [2017/02/10 09:34] – oschuett
Line 38: Line 38:
   * The [[http://www.cp2k.org/static/coverage/coverage_report_combined/src/xc/index.html| test coverage]] of the XC-functionals is pretty low. Since we have libxc as reference, one could easily write a unit-test that compares both implementations by applying them to a randomly generated density.   * The [[http://www.cp2k.org/static/coverage/coverage_report_combined/src/xc/index.html| test coverage]] of the XC-functionals is pretty low. Since we have libxc as reference, one could easily write a unit-test that compares both implementations by applying them to a randomly generated density.
   * Setup a regtester with the [[https://www.pgroup.com/products/community.htm|PGI Community Edition Compiler]].   * Setup a regtester with the [[https://www.pgroup.com/products/community.htm|PGI Community Edition Compiler]].
-  * Performance regression testing. Requires an empty machine for reproducibility. Enumerate important primitives and their possible performance requiems (compute-, communication-, overhead-bound).+  * Performance regression testing. Requires an empty machine for reproducibility. Test each kernel (FFT, LA, grid,...) in different regimes (compute-, communication-, overhead-bound).
  
 ====== Wiki ====== ====== Wiki ======
dev/todo.txt · Last modified: 2020/11/19 17:46 by tmueller