Then the END block I posted handles it.

Sorry, I missed that. Was it a late addition?

I agree that is a better work-around than POSIX::_exit(), but it is still a work-around that should not be necessary.

Think of it this way. strict is optional. warnings are optional. The entire philosophy of Perl is permissive. Why should this one module buck that philosophy and decide that this warning is mandatory?

I don't have a problem with it being on by default--indeed I use strict & warnings in everything I write; and I was a strong advocate for, and may even have had a significant influence in the default-on for these in Perl 6--but the user should be able to supress them through the normal no warnings (or $^W) mechanisms and not have to resort to such work-arounds.

Similarly but even worse in some respects, are the Free to wrong pool messages, which are core developer diagnostics that, for the most part, the user programmer can do nothing about. There is one terse reference in the changes file for Digest::SHA to avoiding this message, but nothing your average application programmer is likely to be able to do much about, never mind the end user. So why do they have to see them?