Path: utzoo!attcan!uunet!lll-winken!lll-lcc!lll-tis!helios.ee.lbl.gov!pasteur!ames!umd5!brl-adm!brl-smoke!gwyn
From: gwyn@brl-smoke.ARPA (Doug Gwyn )
Newsgroups: comp.unix.questions
Subject: Re: AT&T wait(2)
Message-ID: <8218@brl-smoke.ARPA>
Date: 7 Jul 88 22:24:28 GMT
References: <58940@sun.uucp>
Reply-To: gwyn@brl.arpa (Doug Gwyn (VLD/VMB) )
Organization: Ballistic Research Lab (BRL), APG, MD.
Lines: 8

In article <58940@sun.uucp> jpayne%breakpoint@Sun.COM (Jonathan Payne) writes:
>In system V and related systems, is there a version of wait that takes
>flag arguments like wait2() and wait3() in berkeley systems?

Not yet.  There will have to be something soon, for POSIX conformance.
Dave Korn was trying to force wait2() semantics into the wait() system
call (which is already overloaded due to the SIGCLD nonsense), but I
doubt that waitpid() can be emulated reasonably without using a
different system call hook than wait().