IRIX64 6.5, gcc 2.95.1: weird inet_ntoa() behaviour

[David Starks-Browning]

This is a known problem.  You can find discussions and work-arounds
with a Deja-news search of inet_ntoa against comp.sys.sgi.*.

It would be great if gcc could build in a workaround, like a
gcc-version of inet_ntoa for IRIX.

> > Date: Wed, 24 May 2000 10:51:32 -0400 (EDT)
> > From: Denis Vakatov <vakatov@peony.nlm.nih.gov>
> > To: gcc@gcc.gnu.org
> > CC: vakatov@peony.nlm.nih.gov, lavr@peony.nlm.nih.gov
> 
> First, you used the wrong list.  You seem to want to report a bug in
> the compiler.  There is a list just for doing that that we created.
> I'll copy the whole message there.
> 
> Second, you don't provide a complete testcase.  Certails the manual
> and the webpages by now describe what a test case is.
> 
> I suspect the code in the headerfiles is not ANSI C code.  If that is
> the case, we should probably fixincludes it.
> 
> > See the attached code:
> 
> > inet_ntoa() produces an invalid result when the following code
> > compiled with GCC.
> > However, it produces a correct result when compiled with native "cc".