linux-kernel - Re: [RFC] tty: pty -- Add ability to setup next pty number for ptmx open

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:   Thu, 27 Apr 2017 12:20:32 +0300
From:   Cyrill Gorcunov <gorcunov@...il.com>
To:     LKML <linux-kernel@...r.kernel.org>
Cc:     Andrey Vagin <avagin@...nvz.org>,
        Pavel Emelyanov <xemul@...tuozzo.com>,
        "Eric W. Biederman" <ebiederm@...ssion.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Jiri Slaby <jslaby@...e.com>, "H. Peter Anvin" <hpa@...or.com>,
        Andrew Morton <akpm@...uxfoundation.org>
Subject: Re: [RFC] tty: pty -- Add ability to setup next pty number for ptmx open

On Fri, Mar 17, 2017 at 1:56 PM, Cyrill Gorcunov <gorcunov@...il.com> wrote:
> When we restore PTY terminals we do iterate over open() calls to reach
> the index used at checkpoint time (say application opened a number of
> terminals, then closed all other except last one) which of course very
> inconvenient.
>
> The problem get more notable in scope of controlling terminal represented
> by /dev/tty device: current CRIU restore scheme is lazy, we don't open
> files immedeately after the forks but rather defer until the whole
> process tree is created, then every process opens own files in async
> way and in result /dev/tty is not propagated into children because
> it is not yet opened.
>

Gentlemen, ping?