Greg Stein writes:

I'm not very familiar with these APIs, but should you let go of the interpreter lock when you call them? (and for the other new funcs)

None of these should be doing an I/O as far as I can determine. Whenever I get to getlogin() (which AMK & I decided should be included, based on the specs that /F pointed us to), I will release the interpreter lock for the getlogin_r() variant. I'm not sure I should release it for the non-reentrant getlogin(), however; the specification for getlogin*() pretty much requires that it read from utmp. ;( -Fred -- Fred L. Drake, Jr. <fdrake@acm.org> Corporation for National Research Initiatives