linux-kernel - Re: [PATCH] Ban module license tag string termination trick

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:	Thu, 01 Feb 2007 21:19:06 -0500
From:	Valdis.Kletnieks@...edu
To:	Trent Waddington <trent.waddington@...il.com>
Cc:	Tomas Carnecky <tom@...ervice.com>,
	Jon Masters <jonathan@...masters.org>,
	Jan Engelhardt <jengelh@...ux01.gwdg.de>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Andrew Morton <akpm@...l.org>,
	Alexey Dobriyan <adobriyan@...il.com>
Subject: Re: [PATCH] Ban module license tag string termination trick

On Fri, 02 Feb 2007 10:51:23 +1000, Trent Waddington said:
> On 2/2/07, Tomas Carnecky <tom@...ervice.com> wrote:
> > Can't you put this somewhere into the documentation: it's our kernel,
> > play by our rules, and our rules are, the license is what is visible in
> > 'printf(license)'?
> 
> Here I was thinking the rules were: all modules must be GPL and the
> jerks who make proprietary modules are just blatantly breaking the
> law.  But you're right, the MODULE_LICENSE tag really does imply that
> licenses other than the GPL are ok.

Given that the definition of "derived work" in the software world is still
quite squishy and not firmed up, it's not at all clear that proprietary
modules are "blatantly" breaking the law.  In particular, feel free to
cite actual statute or case law that proves unequivocally that modules that
have a GPL shim that interface between the kernel and a binary blob are
breaking the law.

Hint: first you have to prove that the kernel API doesn't qualify as 'scenes
a faire' - a tough job when calling a function in any way but the one approved
one will cause an oops. ;)

Content of type "application/pgp-signature" skipped