Leon Timmermans

> My opinion, requiring pTHX? That can absolutely can be removed. Worst case, a "bug fix" is max 12 months away until next major Perl release. That's because of IMPLICIT_SYS...

I think there are two problems here. One is that you're using 5.30 settings and patches with a 5.38 release, this is causing a bunch of confusion. As a consequence...

> Interesting. My configure flags (based on nixpkgs) are: That looks like you're using perl-cross instead of the usual configure. That may have to be fixed too.

> If this proceeds, I would recommend now using (instead of Crypt::URandom) Crypt::SysRandom which can optionally use the preferred Crypt::SysRandom::XS if that is also installed. Honestly Crypt::SysRandom is [small enough](https://metacpan.org/dist/Crypt-SysRandom/source/lib/Crypt/SysRandom.pm)...

It's hard to say anything about a segmentation fault without a stacktrace, really.

Also just ran into this issue (while building on Arch Linux). Would appreciate this being fixed.

`SV * const dsv` as argument type doesn't really make sense as an argument type. That const doesn't actually affect the interface.

> I checked, and there are about 400 instances in embed.fnc of the same type of declaration Fair. I don't like it but don't let that stop you from merging...

> there is no alternative to attach __DATA__ to package except of old package syntax I get that you might not like that aesthetically, but when is that a practical...

I think most of your comments boil down to "mention the man section when referring to a man page". For our internal documentation, that's a limitation of our pod2man pipeline...