linux-kernel - Re: [PATCH v6 1/2] fork: extend clone3() to support setting a PID

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]

Message-ID: <20190814100402.jf5p2wsqngeuazaj@wittgenstein>
Date:   Wed, 14 Aug 2019 12:04:03 +0200
From:   Christian Brauner <christian.brauner@...ntu.com>
To:     Pavel Emelianov <xemul@...tuozzo.com>,
        Oleg Nesterov <oleg@...hat.com>
Cc:     Adrian Reber <areber@...hat.com>,
        Eric Biederman <ebiederm@...ssion.com>,
        Jann Horn <jannh@...gle.com>,
        Dmitry Safonov <0x7f454c46@...il.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "Andrei Vagin (C)" <avagin@...il.com>,
        Mike Rapoport <rppt@...ux.ibm.com>,
        Radostin Stoyanov <rstoyanov1@...il.com>
Subject: Re: [PATCH v6 1/2] fork: extend clone3() to support setting a PID

On Wed, Aug 14, 2019 at 09:50:03AM +0000, Pavel Emelianov wrote:
> On 8/12/19 11:09 PM, Adrian Reber wrote:
> > The main motivation to add set_tid to clone3() is CRIU.
> > 
> > To restore a process with the same PID/TID CRIU currently uses
> > /proc/sys/kernel/ns_last_pid. It writes the desired (PID - 1) to
> > ns_last_pid and then (quickly) does a clone(). This works most of the
> > time, but it is racy. It is also slow as it requires multiple syscalls.
> > 
> > Extending clone3() to support set_tid makes it possible restore a
> > process using CRIU without accessing /proc/sys/kernel/ns_last_pid and
> > race free (as long as the desired PID/TID is available).
> > 
> > This clone3() extension places the same restrictions (CAP_SYS_ADMIN)
> > on clone3() with set_tid as they are currently in place for ns_last_pid.
> > 
> > Signed-off-by: Adrian Reber <areber@...hat.com>
> 
> Acked-by: Pavel Emelyanov <xemul@...nvz.org>

On Tue, Aug 13, 2019 at 04:30:24PM +0200, Oleg Nesterov wrote:
> On 08/12, Adrian Reber wrote:
> >
> > The main motivation to add set_tid to clone3() is CRIU.
> >
> > To restore a process with the same PID/TID CRIU currently uses
> > /proc/sys/kernel/ns_last_pid. It writes the desired (PID - 1) to
> > ns_last_pid and then (quickly) does a clone(). This works most of the
> > time, but it is racy. It is also slow as it requires multiple syscalls.
> >
> > Extending clone3() to support set_tid makes it possible restore a
> > process using CRIU without accessing /proc/sys/kernel/ns_last_pid and
> > race free (as long as the desired PID/TID is available).
> >
> > This clone3() extension places the same restrictions (CAP_SYS_ADMIN)
> > on clone3() with set_tid as they are currently in place for ns_last_pid.
> >
> > Signed-off-by: Adrian Reber <areber@...hat.com>
> 
> Reviewed-by: Oleg Nesterov <oleg@...hat.com>
> 

Added-to:
https://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux.git/log/?h=pidfd

Merged-into:
https://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux.git/log/?h=for-next

Thanks!
Christian