From owner-freebsd-questions  Wed Jan 18 11:14:49 1995
Return-Path: questions-owner
Received: (from root@localhost) by freefall.cdrom.com (8.6.9/8.6.6) id LAA12206 for questions-outgoing; Wed, 18 Jan 1995 11:14:49 -0800
Received: from sovcom.kiae.su (sovcom.kiae.su [144.206.136.1]) by freefall.cdrom.com (8.6.9/8.6.6) with SMTP id LAA12200; Wed, 18 Jan 1995 11:14:46 -0800
Received: by sovcom.kiae.su id AA21550
  (5.65.kiae-2 ); Wed, 18 Jan 1995 22:04:28 +0300
Received: by sovcom.KIAE.su (UUMAIL/2.0); Wed, 18 Jan 95 22:04:28 +0300
Received: (from ache@localhost) by astral.msk.su (8.6.8/8.6.6) id WAA00619; Wed, 18 Jan 1995 22:00:30 +0300
To: freebsd-current@freefall.cdrom.com, freebsd-questions@freefall.cdrom.com,
        Sean Eric Fagan <sef@kithrup.com>, ugen@netvision.net.il
References: <199501181814.KAA24727@kithrup.com>
In-Reply-To: <199501181814.KAA24727@kithrup.com>;
    from Sean Eric Fagan at Wed, 18 Jan 1995 10:14:19 -0800
Message-Id: <UBDLM7lib2@astral.msk.su>
Organization: Olahm Ha-Yetzirah
Date: Wed, 18 Jan 1995 22:00:29 +0300
X-Mailer: Mail/@ [v2.31 FreeBSD]
From: "Andrew A. Chernov, Black Mage" <ache@astral.msk.su>
X-Class: Fast
Subject: Re: GCC bug again...
Lines: 49
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Length: 1271
Sender: questions-owner@FreeBSD.org
Precedence: bulk

In message <199501181814.KAA24727@kithrup.com> Sean Eric Fagan writes:

>Ah ah.  I see what he's saying.  I compiled:

>extern long long lseek (int, long long, int);

>char buf[1023];
>int fd;

>foo() {
>        lseek (fd, -sizeof(buf), 1);
>}

>with 'gcc -S' on a 1.1++ system, and this was the output:
>        .file   "t.c"
>gcc2_compiled.:
>___gnu_compiled_c:
>.text
>        .align 2
>.globl _foo
>        .type    _foo,@function
>_foo:
>        pushl %ebp
>        movl %esp,%ebp
>        pushl $1
>        pushl $0	# <--- should be -1
>        pushl $-1023
>        pushl _fd
>        call _lseek
>        addl $16,%esp
>L1:
>        leave
>        ret
>Lfe1:
>        .size    _foo,Lfe1-_foo
>.comm _buf,1024
>.comm _fd,4

>If the second argument is cast to '(int)', then it Does The Right Thing.

>gcc-2.6.0 seems to do this, as well, so it is confused about *something*.

It is completely right behaviour, -unsigned undefined and
must be casted to -((off_t)unsigned).
-- 
Andrew A. Chernov        : And I rest so composedly,  /Now, in my bed,
ache@astral.msk.su       : That any beholder  /Might fancy me dead -
FidoNet: 2:5020/230.3    : Might start at beholding me,  /Thinking me dead.
RELCOM Team,FreeBSD Team :         E.A.Poe         From "For Annie" 1849