linux-kernel - Re: /proc/pid/fd && anon_inode

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CA+55aFzbY3zYD1i5x02Z-6K3nHKzJj6227ZBYRLqij9y0eXRqQ@mail.gmail.com>
Date:	Sun, 25 Aug 2013 13:05:55 -0700
From:	Linus Torvalds <torvalds@...ux-foundation.org>
To:	Oleg Nesterov <oleg@...hat.com>
Cc:	Willy Tarreau <w@....eu>, Al Viro <viro@...iv.linux.org.uk>,
	Andy Lutomirski <luto@...capital.net>,
	"security@...nel.org" <security@...nel.org>,
	Ingo Molnar <mingo@...nel.org>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Linux FS Devel <linux-fsdevel@...r.kernel.org>,
	Brad Spengler <spender@...ecurity.net>
Subject: Re: /proc/pid/fd && anon_inode_fops

On Sun, Aug 25, 2013 at 12:48 PM, Oleg Nesterov <oleg@...hat.com> wrote:
>
> Or pid_revalidate(), but my concern is task_dumpable() logic.
>
> pid_revalidate() does inode->i_*id = GLOBAL_ROOT_*ID if task_dumpable()
> fails, but it can fail simply because ->mm = NULL.
>
> This means that almost everything in /proc/zombie-pid/ becomes root.
> Doesn't really hurt, but for what? Looks a bit strange imho.

The zombie case shouldn't be relevant, because a zombie will have
closed all the file descriptors anyway, so they no longer exist.

That said, task_dumpable isn't wonderful, and I suspect we could drop
that logic entirely in the tid-fd case if we just use f_cred.

The reason we have task_dumpable is exactly because we use the task
credentials, and they may not really be relevant to the file
credentials. IOW, it's there to protect against execve'ing a suid
program that opens some protected file and then in setuid()'s back the
the original user after having done the critical stuff.  But
file->f_cred is exactly about the credentials at the time of the open,
so it should make things like that irrelevant.

            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/