From owner-freebsd-arch  Fri Dec 15  6:34: 1 2000
From owner-freebsd-arch@FreeBSD.ORG  Fri Dec 15 06:33:58 2000
Return-Path: <owner-freebsd-arch@FreeBSD.ORG>
Delivered-To: freebsd-arch@freebsd.org
Received: from warning.follo.net (warning.follo.net [195.204.136.30])
	by hub.freebsd.org (Postfix) with ESMTP id C320037B400
	for <arch@FreeBSD.ORG>; Fri, 15 Dec 2000 06:33:48 -0800 (PST)
Received: (from eivind@localhost)
	by warning.follo.net (8.9.3/8.9.3) id PAA77514;
	Fri, 15 Dec 2000 15:33:45 +0100 (CET)
Date: Fri, 15 Dec 2000 15:33:45 +0100
From: Eivind Eklund <eivind@FreeBSD.ORG>
To: Martin Blapp <mb@imp.ch>
Cc: arch@FreeBSD.ORG
Subject: Re: _DIAGASSERT()
Message-ID: <20001215153345.A76398@warning.follo.net>
References: <Pine.BSF.4.21.0012151505110.10655-100000@levais.imp.ch>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.4i
In-Reply-To: <Pine.BSF.4.21.0012151505110.10655-100000@levais.imp.ch>; from mb@imp.ch on Fri, Dec 15, 2000 at 03:08:48PM +0100
Sender: eivind@warning.follo.net
Sender: owner-freebsd-arch@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG

On Fri, Dec 15, 2000 at 03:08:48PM +0100, Martin Blapp wrote:
> 
> Hi,
> 
> I'm porting code from NetBSD and ask myself if I should remove
> some userland-code _DIAGASSERT()'s ...
> 
> There is nothing appropriate in FreeBSD like diagassert() ...
> 
> In NetBSD this is defined therefore :
> 
> include/assert.h:#  define _DIAGASSERT(e) ((e) ? (void)0 
> : __diagassert(__FILE__, __LINE__, "e"))
> 
> and in libc/gen/assert.c :
> 
> void
> __diagassert(file, line, failedexpr)
>         const char *file, *failedexpr;
>         int line;
> {
>                 /*
>                  * XXX: check $DIAGASSERT here, and do user-defined
> actions
>                  */
>         (void)fprintf(stderr,
>             "%s: assertion \"%s\" failed: file \"%s\", line %d\n",
>             __progname, failedexpr, file, line);
>         syslog(LOG_DEBUG|LOG_USER,
>             "assertion \"%s\" failed: file \"%s\", line %d",
>             failedexpr, file, line);
>         return;
> } 

The relevant question is really "Should we add this, and if so in what exact
form - or should we just drop the internal rigidity tests NetBSD has added?"

I think the right answer is to add it, but with LOG_ERR instead of LOG_DEBUG - 
failure of the internal invariants in the code is something we want the user
to see, not syslogd to discard.

Eivind.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-arch" in the body of the message