This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [patch, libstdc++, testsuite] Modify tests to use tmpnam()


On Wed, 2010-03-03 at 14:31 -0800, Steve Ellcey wrote:
> On Wed, 2010-03-03 at 23:15 +0100, Paolo Carlini wrote:
>   
> > Ok. Actually, it seems to me we should use tmpnam more consistently -
> > after all it's a perfectly standard C89 facility - instead of inventing
> > names here and there for temporary files. Can you add a one-line comment
> > about that before the arrays?
> > 
> > Thanks,
> > Paolo.
> 
> 
> I am not sure what kind of comment you want.  Maybe something like:
> 
> // Use tmpnam to create a unique name for the FIFO.
> 
> 
> Hm, I just happened to look at the linux man page for tmpnam (as opposed
> to the HPUX one) and noticed this:
> 
> | BUGS
> |       Never use this function.  Use mkstemp(3) or tmpfile(3) instead.
> 
> But I don't think either of those routines can create a FIFO like we are
> doing in these tests so this seems like a bogus comment in the man page.

I had forgotten that GNU ld warns about use of tmpnam:

   warning: the use of `tmpnam' is dangerous, better use `mkstemp'

It would be messy for the tests to turn off that warning since GNU ld
isn't the only linker used with GCC.  Other tests that use tmpnam are
compile-only and do not link.

Janis


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]