Maetveis wrote: > > This is a lot of tests... do we really need every combination of > > type/operator/flag? I guess it makes sure we don't have gaps, but thousands > > of lines of tests seems excessive. > > It does seem like a lot of testing, but I'd rather we be excessive than take > our historical approach of only getting a fraction of test coverage.
I've reduced the testing somewhat, given that there is no expected difference in basic/improved/full handling for most of `complex-range-real.c` I cut out basic. Also removed most of multiply testing since there should be no difference in modes between them. I think including the full function without optimizations is contributing to the verbosity. The gep/load/store sequences are not really interesting for these tests. I can experiment with `--filter-out`, but I'd prefer to do that as a followup. https://github.com/llvm/llvm-project/pull/131447 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
