linux-kernel - Re: [PATCHv5 01/37] ns: Introduce Time Namespace

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: <4d8d8489-28c8-259f-23a9-ed2b89699b73@gmail.com>
Date:   Fri, 2 Aug 2019 00:46:39 +0100
From:   Dmitry Safonov <0x7f454c46@...il.com>
To:     Andy Lutomirski <luto@...nel.org>, Dmitry Safonov <dima@...sta.com>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        Andrei Vagin <avagin@...nvz.org>,
        Adrian Reber <adrian@...as.de>, Arnd Bergmann <arnd@...db.de>,
        Christian Brauner <christian.brauner@...ntu.com>,
        Cyrill Gorcunov <gorcunov@...nvz.org>,
        "Eric W. Biederman" <ebiederm@...ssion.com>,
        "H. Peter Anvin" <hpa@...or.com>, Ingo Molnar <mingo@...hat.com>,
        Jann Horn <jannh@...gle.com>, Jeff Dike <jdike@...toit.com>,
        Oleg Nesterov <oleg@...hat.com>,
        Pavel Emelyanov <xemul@...tuozzo.com>,
        Shuah Khan <shuah@...nel.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Vincenzo Frascino <vincenzo.frascino@....com>,
        Linux Containers <containers@...ts.linux-foundation.org>,
        criu@...nvz.org, Linux API <linux-api@...r.kernel.org>,
        X86 ML <x86@...nel.org>
Subject: Re: [PATCHv5 01/37] ns: Introduce Time Namespace

Hi Andy,

Thank you for the review,

On 8/1/19 6:29 AM, Andy Lutomirski wrote:
> On Mon, Jul 29, 2019 at 2:58 PM Dmitry Safonov <dima@...sta.com> wrote:
>>
>> From: Andrei Vagin <avagin@...nvz.org>
>>
>> Time Namespace isolates clock values.
> 
>> +static int timens_install(struct nsproxy *nsproxy, struct ns_common *new)
>> +{
>> +       struct time_namespace *ns = to_time_ns(new);
>> +
>> +       if (!thread_group_empty(current))
>> +               return -EINVAL;
> 
> You also need to check for other users of the mm.

Oops. It seems like, if the check was

if (!current_is_single_threaded())
    return -EUSERS;

instead of thread_group_empty(current), it would address the concerns
from 23/37 and 25/37 patches, too?

> 
>> +
>> +       if (!ns_capable(ns->user_ns, CAP_SYS_ADMIN) ||
>> +           !ns_capable(current_user_ns(), CAP_SYS_ADMIN))
>> +               return -EPERM;
>> +
>> +       get_time_ns(ns);
>> +       get_time_ns(ns);
>> +       put_time_ns(nsproxy->time_ns);
>> +       put_time_ns(nsproxy->time_ns_for_children);
>> +       nsproxy->time_ns = ns;
>> +       nsproxy->time_ns_for_children = ns;
>> +       ns->initialized = true;
> 
> I really really wish that setns() took an explicit flag for "change
> now" or "change for children", since the semantics are different.  Oh
> well.
> 

Thanks,
          Dmitry