Fred Moyer wrote: > Gozer, I've refactored everything to use keep alives instead of sameinterp > per your example and our discussion at ApacheCon.
Cool! I saw the commits, but haven't had the chance of testing them. > I haven't tested to see > if it sped up running the test suite, but it should in theory since less > attempts will be made to locate the correct interpreter. And it also removes the non-deterministic aspect of these tests. With the old model, they could sometimes fail to locate the correct interp and effectively not run that particular test. This way is good, IMO, even if things are not any faster, they are more correct. > There is still t/modperl/sameinterp.t and > t/response/TestModperl/sameinterp.pm to implement the actual sameinterp > mechanism. Should I remove those, or is there some part of it we need > that keep_alives doesn't do? If there are no tests left that need them, I'd nuke them. We can always bring them back from the dead if we have a need for it in the future. -- Philippe M. Chiasson GPG: F9BFE0C2480E7680 1AE53631CB32A107 88C3A5A5 http://gozer.ectoplasm.org/ m/gozer\@(apache|cpan|ectoplasm)\.org/
signature.asc
Description: OpenPGP digital signature
