Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
CMake: enable -ffp-exceptions-behavior=strict when compiling with clang
We have to ensure that we emit floating-point instructions in debug mode that preserve the occurence of floating-point exceptions and don't introduce new ones. gcc plays nicely in this regard by enabling `-ftrapping-math` per default, at least for the level of optimization we have in debug mode. clang however is more aggressive and assumes that it can optimize code disregarding precise floating-point exception semantics. We thus set `-ffp-exceptions-behavior=strict` in debug mode to ensure that our testsuite doesn't run into false positive floating-point exceptions. See dealii#15496
- Loading branch information