From owner-freebsd-current  Mon Jun 26  2:18:23 2000
Delivered-To: freebsd-current@freebsd.org
Received: from axl.ops.uunet.co.za (axl.ops.uunet.co.za [196.31.2.163])
	by hub.freebsd.org (Postfix) with ESMTP
	id A42AF37BCE4; Mon, 26 Jun 2000 02:18:15 -0700 (PDT)
	(envelope-from sheldonh@axl.ops.uunet.co.za)
Received: from sheldonh (helo=axl.ops.uunet.co.za)
	by axl.ops.uunet.co.za with local-esmtp (Exim 3.13 #1)
	id 136V1t-000B52-00; Mon, 26 Jun 2000 11:18:09 +0200
From: Sheldon Hearn <sheldonh@uunet.co.za>
To: Stefan Esser <se@FreeBSD.ORG>
Cc: FreeBSD-current@FreeBSD.ORG
Subject: Re: Extend "test" and "expr" to 64 bit integers 
In-reply-to: Your message of "Fri, 23 Jun 2000 19:05:14 +0200."
             <20000623190514.A3633@StefanEsser.FreeBSD.org> 
Date: Mon, 26 Jun 2000 11:18:08 +0200
Message-ID: <42594.962011088@axl.ops.uunet.co.za>
Sender: owner-freebsd-current@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG



On Fri, 23 Jun 2000 19:05:14 +0200, Stefan Esser wrote:

> 1) I choose "quad_t" for long integers. Is this a good choice ?
>    In the kernel I'd use int64_t, but I'm not sure what is most
>    appropriate here.

I'd use the new C standard's int64_t, since you're specifically looking
for 64-bit width integers.

> 2) There is a new dependency on <sys/types.h> in test.c (for any 
>    of our 64 bit integer types).

These exact-width integer types should really be defined in <inttype.h>,
no?

> 3) The changes to "expr" rely on 32 bit integers being promoted 
>    to 64 bit integers in function calls (actually only invocations
>    of make_integer().)

IT's probably worth chatting the the NetBSD folks about this, since
that's where we got our version of test(1).

> What does Posix say about these programs ?

As long as your operands are integers, POSIX.2 is happy.

Ciao,
Sheldon.


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