linux-kernel - Re: fs/proc/base.c: text md5sums; tgid vs tid; and INF vs ONE?

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: <87ip9refzk.fsf@xmission.com>
Date:	Tue, 30 Oct 2012 15:45:35 -0700
From:	ebiederm@...ssion.com (Eric W. Biederman)
To:	Arvid Brodin <Arvid.Brodin@...n.com>
Cc:	"linux-kernel\@vger.kernel.org" <linux-kernel@...r.kernel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Al Viro <viro@...iv.linux.org.uk>,
	Cyrill Gorcunov <gorcunov@...nvz.org>,
	"David Rientjes" <rientjes@...gle.com>
Subject: Re: fs/proc/base.c: text md5sums; tgid vs tid; and INF vs ONE?

Arvid Brodin <Arvid.Brodin@...n.com> writes:

> Hi,
>
> Below is a patch that adds a file /proc/PID/text_md5sum which when read returns the md5
> checksum of a process' text segment. (This would be used e.g. to make sure a process'
> code hasn't been tampered with.)
>
> However, I have a few questions:
>
> * What's the difference between the tgid_base_stuff and tid_base_stuff arrays? (One for
> processes and one for the process' threads? I haven't been able to find any info about
> this so I'm guessing.)

Yes.  One for thread groups and one for threads.

> * When should I use the INF ("read") vs the ONE ("show") macro?

proc_read depends on the caller to allocate a 4k buffer, instead of
sizing the buffer based upon the size of the text being written.  Which
makes proc_read an error prone and ultimately deprecated way of handling
things.

Using some variant on seq_file is preferred for new files.

> * Any other comments about the code?

There are known successful attacks against md5 so using md5 for
something new and security related is a bad idea.

Userspace can just as easily compute a security hash itself you don't
need kernel support.

I recommend you checkout the code in security/ima/ looks like it can
already do what you are trying to do.

Eric
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/