Relay-Version: version B 2.10.2 9/18/84; site lsuc.UUCP
Posting-Version: version B 2.10.1 6/24/83; site itm.UUCP
Path: lsuc!pesnta!itm!danny
From: danny@itm.UUCP (Danny)
Newsgroups: pe.cust.bugs
Subject: Re: what does wait(2) really do with status?
Message-ID: <225@itm.UUCP>
Date: 8 Jan 85 16:09:52 GMT
Date-Received: 9 Jan 85 08:18:22 GMT
References: <268@lsuc.UUCP>
Reply-To: danny@itm.UUCP (Danny)
Organization: In Touch - Atlanta, GA
Lines: 31

David Sherman writes:
    >According to the manual for wait(2), the argument supplied
    >is a pointer to int (that is, "int *status"). The manual goes
    >on to talk about its "high byte" and "low byte", obviously
    >a throwback to the PDP-11. Anyone know whether status should
    >be declared pointer to short, or what?

    This is the way I've done it, it's simple, it requires no
masking/shifting, etc.

/*
 *      Structure of return from wait() system call
 *      Note:  This is probably machine dependent
 */
struct status {
        char    dummy1; /* dummy char # 1 */
        char    dummy2; /* dummy char # 2 */
        char    ret;    /* Exit code from process */
        char    pstat;  /* Signal process died with (if any) */
};

    The note above is correct, it is very machine-dependant.  The
VAX version reverses pstat and ret, and defines them first in the
struct (I think).

    Anyway, try it out...

                                        Danny
-- 
				Daniel S. Cox
				({gatech|akgua}!itm!danny)