Genrerate the specialization files by grouping the specializations into
five groups, cutting build time by three minutes (roughly 20%) in my
tests.
Change-Id: Ifd6a4523a47346c5e84c1cda15b7abba60d8e953
The initial reason for this is because of a previous reformatting CL
triggered a macro redefinition warning in the schur eliminator. But
actually it was worse because the reordering had caused the macro
definition to be ignored and caused a performance regression.
This simplifies the generated files, fixes some formatting errors
and recovers the performance.
Change-Id: I9dbeffc38743b3f24b25843feec2e26a73188413
- update file generator scripts / templates so generated files adhere
to clang-format
- A few exceptions are not fixed, where the file generation results in
lines of different width. To properly fix this would make the code
more complicated and it's not that important for generated files
anyway.
- note that generated files are excluded in ./scripts/format_all.sh
Change-Id: I4f42c83d1fec01242eada5e7ce6c1a5192234d37
1. Refactor the python code that generates the template specializations
to remove code duplication.
2. Improved the logic for template specialization selection where
Eigen::Dynamic now serves as a wildcard.
3. Added schur_templates.h/cc which allows querying the set of available
template specializations without instantiating a linear solver.
4. Added Solver::Summary::schur_structre_given and
Solver::Summary::schur_structure_used and expose them in
Solver::Summary::FullReport for better performance debugging.
5. Updated the templates with newer dates and some minor comments cleanup
which lead to the the template specializations to be re-generated.
Change-Id: Iaf3c6f714353597899916c300465da01f151c3de
Sameer Agarwal