linux-kernel - Re: [PATCH] kernel/cred: Fix bug on cred reference count

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]

Message-ID: <10155.1310202827@redhat.com>
Date:	Sat, 09 Jul 2011 10:13:47 +0100
From:	David Howells <dhowells@...hat.com>
To:	Edwin van Vliet <edwin@...atah.nl>
Cc:	dhowells@...hat.com, akpm@...ux-foundation.org,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] kernel/cred: Fix bug on cred reference count

Edwin van Vliet <edwin@...atah.nl> wrote:

> The comment above states that the old obj ref and subj ref should BOTH be
> released. However, the line is repeated so the old obj ref is released twice
> yet the subj ref is not. So either the comment is wrong, or this is simpy a
> bug. Releasing the same ref twice seems wrong in any case.

The comment is correct and the double release is correct.

task_struct has *two* refs to the cred being replaced and both need to be
updated:

	current->real_cred	- The objective cred
	current->cred		- The subjective cred

The subjective cred is the one used by a task to access other objects; the
objective cred details how the task is perceived by the outside world and
affects how other tasks act upon it.

Normally a task's two cred pointers point to the same cred, but under some
circumstances (such as in cachefiles), the subjective cred can be overridden.

David
--
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/