[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CA+55aFzv3X3yW8WZGMZO+aKa=qWi2w8=zn2DP-Q3jRry8Ermyw@mail.gmail.com>
Date: Tue, 21 Feb 2012 19:47:38 -0800
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: David Miller <davem@...emloft.net>, Ian Kent <raven@...maw.net>
Cc: linux-kernel@...r.kernel.org, hpa@...or.com,
autofs@...ux.kernel.org, thomas@...3r.de, viro@...iv.linux.org.uk
Subject: Re: compat: autofs v5 packet size ambiguity - update
On Tue, Feb 21, 2012 at 7:33 PM, Linus Torvalds
<torvalds@...ux-foundation.org> wrote:
>
> But I'm sure that the autofs daemon does something simple that we can
> depend on in practice.
I didn't find the most recent source code, but the first thing that
automount.c does after calling "mount" (which is a for+exec, so we
cannot take the mounting is_compat_task() into account is to open the
new root for the ioctlfd, and then it does a stat.
And then it does a AUTOFS_IOC_PROTOVER ioctl to see what the protocol
version is.
So we could just decide that
(a) we add a mount option for the packet size (or just "v6" - which
would be "v5 with a fixed packet size")
(b) in the absence of an explicit mount option, we look at
is_compat_task() for the first AUTOFS_IOC_PROTOVER ioctl we get.
That looks fairly straightforward and safe. Hmm?
Where are the automount sources supposed to be, anyway? kernel.org has
a v5 directory, but it's empty.
Linus
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists