Bash 2.05 'configure' unnecessary double-check for 'wait3'

[Paul Eggert]

When I configure Bash 2.05, it outputs this:

checking for wait3 that fills in rusage... no
...
checking for wait3... yes

The second check overrides the result of the first check, so the first
check is redundant.  I looked at the code, and Bash never needs the
rusage part of wait3, so the first check is redundant anyway.

Here is a patch.

2001-06-13  Paul Eggert  <eggert@twinsun.com>

        * configure.in (AC_FUNC_WAIT3): Remove: it was not used, since a later
        `AC_CHECK_FUNC(wait3, AC_DEFINE(HAVE_WAIT3))' overrode the result.
        Bash doesn't need the special feature of AC_FUNC_WAIT3 anyway.

===================================================================
RCS file: configure.in,v
retrieving revision 2.5.0.3
retrieving revision 2.5.0.4
diff -pu -r2.5.0.3 -r2.5.0.4
--- configure.in        2001/05/01 01:00:09     2.5.0.3
+++ configure.in        2001/06/13 15:55:56     2.5.0.4
@@ -446,7 +446,6 @@ AC_FUNC_ALLOCA
 AC_FUNC_GETPGRP
 AC_FUNC_SETVBUF_REVERSED
 AC_FUNC_VPRINTF
-AC_FUNC_WAIT3
 AC_FUNC_STRCOLL
 
 dnl if vprintf is not in libc, see if it's defined in stdio.h