[netperf-talk] OK, this time I really mean it?-)

[Rick Jones]

Andrew Gallatin wrote:
> 
>> Is there a #define that is consistently set when compiling on Solaris?  
> 
> 
> The rest of the code seems to use __sun.  I just tried this, and it
> seems to work.
> 
> Drew
> 
> 
> ------------------------------------------------------------------------
> 
> Index: nettest_bsd.c
> ===================================================================
> --- nettest_bsd.c	(revision 301)
> +++ nettest_bsd.c	(working copy)
> @@ -647,6 +647,11 @@
>    int    change_info = 0;
>    static int change_warning_displayed = 0;
>  
> +#ifdef __sun
> +  if (!debug)	
> +    change_warning_displayed = -1;
> +#endif
> +
>    int count = 0;
>    int error = 0;
>  

OK, try this then - it uses #ifdef __sun and restricts itself to the kludge for 
the bug in question rather than all warnings:

raj at tardy:~/netperf2_trunk$ svn diff src/nettest_bsd.c
Index: src/nettest_bsd.c
===================================================================
--- src/nettest_bsd.c   (revision 301)
+++ src/nettest_bsd.c   (working copy)
@@ -813,8 +813,15 @@
    /* if we dropped the protocol hint, it would be for a protocol that
       getaddrinfo() wasn't supporting yet, not for the bug that it took
       our hint and still returned zero. raj 2006-10-16 */
+  /* as there is now an open bug against (Open)Solaris on this
+     behaviour we will only emit this warning if debug is set under
+     Solaris and will continue to emit it under any circumstance on
+     other platforms should it arise. raj 2009-06-03 */
    if ((change_info & CHANGED_PROTOCOL) &&
        !(change_warning_displayed & CHANGED_PROTOCOL) &&
+#ifdef __sun
+      (debug) &&
+#endif
        (hints.ai_protocol != 0)) {
      change_warning_displayed |= CHANGED_PROTOCOL;
      fprintf(where,