linux-kernel - Re: [PATCH v1 20/22] Move common usercopy into security_getpeersec

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 3 Aug 2018 08:27:47 -0700
From:   Casey Schaufler <casey@...aufler-ca.com>
To:     Piotr Sawicki <p.sawicki2@...tner.samsung.com>,
        LSM <linux-security-module@...r.kernel.org>,
        LKLM <linux-kernel@...r.kernel.org>,
        Paul Moore <paul@...l-moore.com>,
        Stephen Smalley <sds@...ho.nsa.gov>,
        SE Linux <selinux@...ho.nsa.gov>,
        "SMACK-discuss@...ts.01.org" <SMACK-discuss@...ts.01.org>,
        John Johansen <john.johansen@...onical.com>,
        Kees Cook <keescook@...omium.org>,
        Tetsuo Handa <penguin-kernel@...ove.sakura.ne.jp>,
        James Morris <jmorris@...ei.org>
Cc:     "Schaufler, Casey" <casey.schaufler@...el.com>,
        Casey Schaufler <casey@...aufler-ca.com>
Subject: Re: [PATCH v1 20/22] Move common usercopy into
 security_getpeersec_stream

On 8/3/2018 2:10 AM, Piotr Sawicki wrote:
> On 07/16/2018 08:24 PM, Casey Schaufler wrote:
>> [PATCH 20/22] Move common usercopy into security_getpeersec_stream
>>
>> The modules implementing hook for getpeersec_stream
>> don't need to be duplicating the copy-to-user checks.
>> Moving the user copy part into the infrastructure makes
>> the security module code simpler and reduces the places
>> where user copy code may go awry.
> Hi,
>
> This change will break the API. Some clients may call getsockopt(..,SO_PEERSEC,..) twice. Firstly, to fetch the length. In that case xxx_socket_getpeersec_stream should return -ERANGE and set *optlen. Secondly, to retrieve a proper security label.

Nuts. You're correct. I will fix this in the next round.

>
> Please take a look at the implementation of the getClientSmackLabel() function in Cynara: https://review.tizen.org/gerrit/#/c/26888/6/src/helpers/creds-socket/creds-socket-inner.cpp
> Also there is an email thread about this socket option which tells us why it is made this way: "[RFC] SO_PEERSEC - security credentials for Unix stream sockets"
>
> http://lists.jammed.com/linux-security-module/2003/12/0029.html
>
> Regards,
> Piotr
>