linux-kernel - Re: [RFC][PATCH 2/2] binfmt_elf: FatELF support for kernel modules.

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]

Message-ID: <alpine.OSX.1.10.0910232026080.2186@caridad.icculuslan>
Date:	Fri, 23 Oct 2009 20:31:16 -0400 (EDT)
From:	"Ryan C. Gordon" <icculus@...ulus.org>
To:	Jeremy Fitzhardinge <jeremy@...p.org>
cc:	linux-kernel@...r.kernel.org
Subject: Re: [RFC][PATCH 2/2] binfmt_elf: FatELF support for kernel
 modules.

On Fri, 23 Oct 2009, Jeremy Fitzhardinge wrote:

> Any open source driver should be encouraged to be merged with mainline
> Linux so there's no need to distribute them separately.  With the
> staging/ tree, that's easier than ever.

Yes, absolutely. I agree.

But even for mainline drivers, there's a compelling use-case for this, 
even for a Linux distribution that's 100% Free Software.

Those that want to play outside the sandbox still have all the misery of 
having to keep up with kernel interface changes and supporting multiple 
kernel releases. That's the hard part for them, and I'm happy to continue 
to make it uncomfortable in that regard. How the kernel module gets loaded 
is probably not of great importance compared to the engineering resources 
they probably devote to just keeping up.

> Well, ideally a fat module would allow modules for multiple kernels to
> be bundled together (same and/or different architectures), which is
> primarily useful for 3rd-party binary distributions.

If the OSABI version doesn't change, they can't use FatELF to pack 
together multiple drivers for different kernel versions. fatelf-glue 
refuses to merge them, and the kernel will pick the first one, and if it 
fails to load, tough luck.

I think this should resolve the concern from a technical standpoint. They 
still have the same problems.

--ryan.

--
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/