linux-kernel - Re: [RFC] module: Strict per-modname namespaces

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: <Zyv-yxClglfwvmUa@bombadil.infradead.org>
Date: Wed, 6 Nov 2024 15:42:03 -0800
From: Luis Chamberlain <mcgrof@...nel.org>
To: Peter Zijlstra <peterz@...radead.org>
Cc: x86@...nel.org, hpa@...or.com, petr.pavlu@...e.com,
	samitolvanen@...gle.com, da.gomez@...sung.com, masahiroy@...nel.org,
	nathan@...nel.org, nicolas@...sle.eu, linux-kernel@...r.kernel.org,
	linux-modules@...r.kernel.org, linux-kbuild@...r.kernel.org,
	hch@...radead.org, gregkh@...uxfoundation.org
Subject: Re: [RFC] module: Strict per-modname namespaces

On Wed, Nov 06, 2024 at 08:02:40PM +0100, Peter Zijlstra wrote:
> Hi,
> 
> I've been wanting $topic for a while, and having just stumbled into the
> whole namespace thing by accident, I figured I'd give it a go, most if
> the hard parts seem to have already been done.

Neat, hch suggested something similar a while back

https://lore.kernel.org/all/ZCIiBHyrzDoTJPXT@bombadil.infradead.org/

> It reserves and disallows imports on any "MODULE_${name}" namespace,
> while it implicitly adds the same namespace to every module.

That's simple enough, I like it.

> This allows exports targeted at specific modules and no others -- one
> random example included. I've hated the various kvm exports we've had
> for a while, and strictly limiting them to the kvm module helps
> alleviate some abuse potential.

Yeah we also want:

EXPORT_SYMBOL_NS_GPL(bdev_disk_changed, MODULE_loop);
EXPORT_SYMBOL_NS_GPL(bdev_disk_changed, MODULE_dasd_kmod);

But we might as well have EXPORT_SYMBOL_GPL_FOR() with the implied
module list. We could then add just:


EXPORT_SYMBOL_GPL_FOR(bdev_disk_changed, loop);

But it would be nice to just also support this as well:

EXPORT_SYMBOL_GPL_FOR(bdev_disk_changed, loop, dasd_kmod);

That's possible perhaps something grotesque like this:

#define _EXPORT_SYMBOL_GPL_FOR_EACH(name, module) EXPORT_SYMBOL_NS_GPL(name, MODULE_##module);

#define _GET_MACRO(_1, _2, _3, _4, _5, NAME, ...) NAME

#define _EXPORT_SYMBOL_GPL_FOR1(name, a) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, a)

#define _EXPORT_SYMBOL_GPL_FOR2(name, a, b) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, a) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, b)

#define _EXPORT_SYMBOL_GPL_FOR3(name, a, b, c) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, a) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, b) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, c)

#define _EXPORT_SYMBOL_GPL_FOR4(name, a, b, c, d) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, a) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, b) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, c) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, d)

#define _EXPORT_SYMBOL_GPL_FOR5(name, a, b, c, d, e) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, a) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, b) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, c) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, d) \
    _EXPORT_SYMBOL_GPL_FOR_EACH(name, e)

#define EXPORT_SYMBOL_GPL_FOR(name, ...) \
    _GET_MACRO(__VA_ARGS__, \
               _EXPORT_SYMBOL_GPL_FOR5, \
               _EXPORT_SYMBOL_GPL_FOR4, \
               _EXPORT_SYMBOL_GPL_FOR3, \
               _EXPORT_SYMBOL_GPL_FOR2, \
               _EXPORT_SYMBOL_GPL_FOR1)(name, __VA_ARGS__)

  Luis