This is the mail archive of the
mailing list for the glibc project.
Re: tmpnam v mkstemp
> Date: Sat, 11 Nov 2000 23:11:14 +0000 (GMT)
> From: "Joseph S. Myers" <firstname.lastname@example.org>
> > Also, POSIX does not allow the warning.
> Clause, subclause and line numbers? ISO/IEC 9945-2:1993(E) ...
Sorry -- I thought that the warning was at run-time, when the program
starts up. If the warning is only at link-time, then of course you're
correct. A conforming compiler is allowed to generate whatever
diagnostics it likes, including false alarms.
Still, it would be nice to be able to disable alarms like this
conveniently, if you know them to be false.