linux-kernel - Re: [PATCH v3 2/4] pid: Add PIDFD_IOCTL_GETFD to fetch file descriptors from processes

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] [day] [month] [year] [list]

Date:   Tue, 17 Dec 2019 12:19:27 +0100
From:   Christian Brauner <christian.brauner@...ntu.com>
To:     Arnd Bergmann <arnd@...db.de>
Cc:     Sargun Dhillon <sargun@...gun.me>,
        LKML <linux-kernel@...r.kernel.org>,
        Linux Containers <containers@...ts.linux-foundation.org>,
        Linux API <linux-api@...r.kernel.org>,
        Linux FS-devel Mailing List <linux-fsdevel@...r.kernel.org>,
        Tycho Andersen <tycho@...ho.ws>, Jann Horn <jannh@...gle.com>,
        Aleksa Sarai <cyphar@...har.com>,
        Oleg Nesterov <oleg@...hat.com>,
        Andy Lutomirski <luto@...capital.net>,
        Al Viro <viro@...iv.linux.org.uk>, gpascutto@...illa.com,
        ealvarez@...illa.com, Florian Weimer <fweimer@...hat.com>,
        jld@...illa.com
Subject: Re: [PATCH v3 2/4] pid: Add PIDFD_IOCTL_GETFD to fetch file
 descriptors from processes

On Tue, Dec 17, 2019 at 09:54:40AM +0100, Arnd Bergmann wrote:
> On Tue, Dec 17, 2019 at 3:50 AM Sargun Dhillon <sargun@...gun.me> wrote:
> > On Mon, Dec 16, 2019 at 5:50 PM Christian Brauner <christian.brauner@...ntu.com> wrote:
> Finally, there is the question whether this should be an ioctl
> operation at all, or
> if it would better be done as a proper syscall. Functionally the two
> are the same
> here, but doing such a fundamental operation as an ioctl doesn't feel
> quite right
> to me. As a system call, this could be something like
> 
> int pidfd_get_fd(int pidfd, int their_fd, int flags);
> 
> along the lines of dup3().

Thanks for taking a look, Arnd!

Yeah, Oleg hinted at this in the first version as well. I originally
disagreed but we can sure also do this as a separate syscall.
What we should keep in mind is that people already brought up adding new
fds to a task. Which is not a problem just something to remember as it
might potentially mean another syscall.

Christian