From xemacs-m  Sat Dec 28 14:13:24 1996
Received: from venus.Sun.COM (venus.Sun.COM [192.9.25.5])
          by xemacs.cs.uiuc.edu (8.8.4/8.8.4) with SMTP
	  id OAA05947 for <xemacs-beta@xemacs.org>; Sat, 28 Dec 1996 14:13:23 -0600 (CST)
Received: from Eng.Sun.COM ([129.146.1.25]) by venus.Sun.COM (SMI-8.6/mail.byaddr) with SMTP id MAA29936; Sat, 28 Dec 1996 12:12:50 -0800
Received: from kindra.eng.sun.com by Eng.Sun.COM (SMI-8.6/SMI-5.3)
	id MAA14925; Sat, 28 Dec 1996 12:12:50 -0800
Received: from xemacs.eng.sun.com by kindra.eng.sun.com (SMI-8.6/SMI-SVR4)
	id MAA19709; Sat, 28 Dec 1996 12:12:47 -0800
Received: by xemacs.eng.sun.com (SMI-8.6/SMI-SVR4)
	id MAA05102; Sat, 28 Dec 1996 12:12:47 -0800
Date: Sat, 28 Dec 1996 12:12:47 -0800
Message-Id: <199612282012.MAA05102@xemacs.eng.sun.com>
From: Martin Buchholz <mrb@Eng.Sun.COM>
To: Steven L Baur <steve@miranova.com>,
        Axel Seibert <aseibert@cybernet-ag.net>
Cc: xemacs-beta@xemacs.org
Subject: Re: xemacs-19.15b4/b4 report for NEXTSTEP
In-Reply-To: <m2681mbhif.fsf@altair.xemacs.org>
References: <199612281122.MAA18514@sol1.cybernet-ag.net>
	<m2681mbhif.fsf@altair.xemacs.org>
Reply-To: Martin Buchholz <mrb@Eng.Sun.COM>
Mime-Version: 1.0 (generated by tm-edit 7.97)
Content-Type: text/plain; charset=US-ASCII

>>>>> "Steven" == Steven L Baur <steve@miranova.com> writes:

>>>>> "Axel" == Axel Seibert <aseibert@cybernet-ag.net> writes:
Axel> typedef SIGTYPE (*signal_handler_t) (int);

Martin> Looks harmless enough. Please dig some more. Often examining
Martin> the source files after it's been pre-processed is useful.

Steven> Here is the patch that was applied to beta4:

Steven> Index: syssignal.h
Steven> ===================================================================
Steven> RCS file: /usr/local/xemacs/xemacs-19.15/src/syssignal.h,v
Steven> retrieving revision 1.1.1.1
Steven> retrieving revision 1.2
Steven> diff -u -r1.1.1.1 -r1.2
Steven> --- syssignal.h 1996/12/18 03:38:24     1.1.1.1
Steven> +++ syssignal.h 1996/12/22 00:29:52     1.2
Steven> @@ -88,11 +88,12 @@
Steven>     the risk of someone forgetting this convention and calling
Steven>     signal() directly. */
 
Steven> +typedef SIGTYPE (*signal_handler_t) (int);
Steven> +
Steven>  #if defined (HAVE_SIGPROCMASK)
 
Steven>  /* The POSIX way (sigaction, sigprocmask, sigpending, sigsuspend) */
 
Steven> -typedef SIGTYPE (*signal_handler_t) (int);
Steven>  extern signal_handler_t sys_do_signal (int signal_number,
Steven>                                        signal_handler_t action);
Steven>  /* Provide our own version of signal(), that calls sigaction().  The

OK, I vaguely remember doing that now, as part of my quest for
warning-free compiles.  It's a non-functional change, and can be
undone if we want to.

The line:

#define signal_handler_t int

in next.h and nextstep.h looks very suspect - signal_handler_t is
supposed to be a function type, not an integral type.

Axel, perhaps you want to take another look at the header files.  Type
correctness is worth it in the long run.  If you really need to have
this change undone, we'll make it so.

Steven> That typedef was moved outside the `#if defined (HAVE_SIGPROCMASK)' guard.
Steven> If it's choking and dying on NeXT, give me a NeXT-specific guard
Steven> symbol to wrap around it.

Martin

