linux-kernel - Re: firmware: dmi-sysfs: why is the access mode of dmi sysfs entries restricted to 0400?

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]

Date:   Wed, 4 Dec 2019 17:01:06 +0800
From:   Guoheyi <guoheyi@...wei.com>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>
CC:     Mike Waychison <mikew@...gle.com>, <linux-kernel@...r.kernel.org>,
        wanghaibin 00208455 <wanghaibin.wang@...wei.com>,
        Thomas Gleixner <tglx@...utronix.de>
Subject: Re: firmware: dmi-sysfs: why is the access mode of dmi sysfs entries
 restricted to 0400?


在 2019/12/4 15:41, Greg Kroah-Hartman 写道:
> On Wed, Dec 04, 2019 at 03:31:22PM +0800, Guoheyi wrote:
>> Hi,
>>
>> Why is the access mode of dmi sysfs entries restricted to 0400? Is it for
>> security concern? If it is, which information do we consider as privacy?
> There's lots of "interesting" information in dmi entries that you
> probably do not want all processes reading, which is why they are
> restricted.
>
>> We would like to fetch CPU information from non-root application, is there
>> feasible way to do that?
> What specific CPU information is not currently exported in /proc/cpuinfo
> that only shows up in DMI entries that you are interested in?

We'd like to get processor manufacturer, speed and version, and pass the 
information to qemu virtual machine, for users of VM might be happy to 
see this instead of "unknown xxx", while qemu may run as non-root.

>
> You can always have root change the permissions of a sysfs file if you
> have a service that wants to allow non-root programs to read specific
> entries.

Thanks; we'll try it.

Heyi

>
> thanks,
>
> greg k-h
>
> .