Possible *small* memory leak?

[ghost]

While testing my program for memory leaks with valgrind, i happened to stumble upon 4 bytes that plibsys allocates not being freed.... while 4 bytes doesn't hugely matter, seeing as plibsys' goal is to run on embedded systems, possibly there it could mean a bit more? Image of the valgrind output, which also states where in the library you might find the memory leak.

[ghost]

I will try the git master version, currently using latest release version, which is 0.4
Found out what might be the reason... very quick and simple fix if the git master version also still has the issue

[ghost]

Fix... all i really did add was 2 extra lines at the bottom...

Also, now no memory leaks reported by valgrind...

[ghost]

Seems like master still has the "bug"

like i said, doesn't hugely matter, and im just glad you even made the library! Good job for almost leak free portable library, it really does make my life easier when i want to port my applications to other operating systems.

[TheRealMichaelWang]

I don't really think this is a bug. There's a puthread_replace_local(or some function like that) that actually calls the destructor. I don't think calling the behavior is intended behavior for set local.