[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CA+55aFwvgY9DLPu3=gtH-kOQ7M-zd94tL60Ts+WB6oNg3c2CKg@mail.gmail.com>
Date: Tue, 11 Dec 2012 12:31:15 -0800
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Eric Paris <eparis@...isplace.org>
Cc: Mimi Zohar <zohar@...ux.vnet.ibm.com>,
"Kasatkin, Dmitry" <dmitry.kasatkin@...el.com>,
Al Viro <viro@...iv.linux.org.uk>,
linux-fsdevel <linux-fsdevel@...r.kernel.org>,
LSM List <linux-security-module@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
James Morris <jmorris@...ei.org>
Subject: Re: [PATCH 0/2] ima: policy search speedup
On Tue, Dec 11, 2012 at 12:15 PM, Eric Paris <eparis@...isplace.org> wrote:
>
> I know it isn't relevant to the final solution, but this is simply
> wrong. IS_PRIVATE() means 'this inode is filesystem internal.' It is
> not used by anything except rieser and the anon_inode. If it is used
> by psuedo filesystems in general, like /proc, shmem mappings, and
> pipes that is a huge bug and is absolutely wrong.
Hmm.. The magic anonfs inode definitely sets it, as far as I can tell.
But it turns out that I was wrong anyway, and you are largely right:
pipes and sockets don't use the anonfs inode (they allocate their own
inodes directly using 'new_inode_pseudo()'), and neither does /proc.
So it's actually only signalfd and timerfd and some other special
things like kvm internal file descriptors that use it.
So never mind about S_PRIVATE, it has odd semantics.
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