From owner-freebsd-hackers  Sun Apr 21 12:25: 6 2002
Delivered-To: freebsd-hackers@freebsd.org
Received: from mailhub.yumyumyum.org (dsl092-171-091.wdc1.dsl.speakeasy.net [66.92.171.91])
	by hub.freebsd.org (Postfix) with SMTP id 6277C37B419
	for <freebsd-hackers@freebsd.org>; Sun, 21 Apr 2002 12:24:54 -0700 (PDT)
Received: (qmail 16373 invoked from network); 21 Apr 2002 19:24:44 -0000
Received: from ken.yumyumyum.org (192.168.0.2)
  by router.yumyumyum.org with SMTP; 21 Apr 2002 19:24:44 -0000
Content-Type: text/plain;
  charset="us-ascii"
From: Kenneth Culver <culverk@yumyumyum.org>
To: freebsd-hackers@freebsd.org, freebsd-emulation@freebsd.org
Subject: implementing linux mmap2 syscall
Date: Sun, 21 Apr 2002 15:25:08 -0400
X-Mailer: KMail [version 1.4]
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Message-Id: <200204211525.08827.culverk@yumyumyum.org>
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
List-ID: <freebsd-hackers.FreeBSD.ORG>
List-Archive: <http://docs.freebsd.org/mail/> (Web Archive)
List-Help: <mailto:majordomo@FreeBSD.ORG?subject=help> (List Instructions)
List-Subscribe: <mailto:majordomo@FreeBSD.ORG?subject=subscribe%20freebsd-hackers>
List-Unsubscribe: <mailto:majordomo@FreeBSD.ORG?subject=unsubscribe%20freebsd-hackers>
X-Loop: FreeBSD.ORG

Hi,
=09I have recently been trying to implement the linux mmap2 syscall into =
our=20
linuxulator, and I have run into a little problem.=20

I looked at the code that was used to implement the regular linux_mmap=20
syscall, and I've also looked in the linux kernel at the code that they u=
se=20
for mmap and mmap2. Basically this is in the linux kernel:

This is what mmap does in linux...

static inline unsigned long do_mmap(struct file *file, unsigned long addr=
,
=09unsigned long len, unsigned long prot,
=09unsigned long flag, unsigned long offset)
{
=09unsigned long ret =3D -EINVAL;
=09if ((offset + PAGE_ALIGN(len)) < offset)
=09=09goto out;
=09if (!(offset & ~PAGE_MASK))
=09=09ret =3D do_mmap_pgoff(file, addr, len, prot, flag, offset >> PAGE_S=
HIFT);
out:
=09return ret;
}

This is what mmap2 does:

andstatic inline long do_mmap2(
=09unsigned long addr, unsigned long len,
=09unsigned long prot, unsigned long flags,
=09unsigned long fd, unsigned long pgoff)
{
=09int error =3D -EBADF;
=09struct file * file =3D NULL;

=09flags &=3D ~(MAP_EXECUTABLE | MAP_DENYWRITE);
=09if (!(flags & MAP_ANONYMOUS)) {
=09=09file =3D fget(fd);
=09=09if (!file)
=09=09=09goto out;
=09}

=09down_write(&current->mm->mmap_sem);
=09error =3D do_mmap_pgoff(file, addr, len, prot, flags, pgoff);
=09up_write(&current->mm->mmap_sem);

=09if (file)
=09=09fput(file);
out:
=09return error;
}

So what it looks like to me is that mmap2 expects an offset that's alread=
y=20
page-aligned (I'm not sure if this is the right way to say it), where mma=
p=20
doesn't. the FreeBSD code in the linuxulator basically just takes the off=
set=20
that is passed in with the linux mmap, and uses that to call FreeBSD's mm=
ap=20
(the kernel version, not the one called from userland). So basically I'm=20
kinda stuck as to what to do to implement linux's mmap2. The only thing I=
 can=20
think of is to implement a FreeBSD "mmap2" that basically assumes that th=
e=20
offset passed in is already page aligned or whatever, and just uses it, a=
nd=20
then have linux_mmap2() just call the FreeBSD mmap2(). Any ideas?

Ken

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