----8<----
But one gray area in particular is something like a driver that was originally written for another operating system (ie clearly not a derived work of Linux in origin). At exactly what point does it become a derived work of the kernel (and thus fall under the GPL)?
THAT is a gray area, and _that_ is the area where I personally believe that some modules may be considered to not be derived works simply because they weren't designed for Linux and don't depend on any special Linux behaviour.
Basically:
- anything that was written with Linux in mind (whether it then _also_ works on other operating systems or not) is clearly partially a derived work.
- anything that has knowledge of and plays with fundamental internal Linux behaviour is clearly a derived work. If you need to muck around with core code, you're derived, no question about it.
Historically, there's been things like the original Andrew filesystem module: a standard filesystem that really wasn't written for Linux in the first place, and just implements a UNIX filesystem. Is that derived just because it got ported to Linux that had a reasonably similar VFS interface to what other UNIXes did? Personally, I didn't feel that I could make that judgment call. Maybe it was, maybe it wasn't, but it clearly is a gray area.
Personally, I think that case wasn't a derived work, and I was willing to tell the AFS guys so.
Does that mean that any kernel module is automatically not a derived work? HELL NO! It has nothing to do with modules per se, except that non-modules clearly are derived works (if they are so central to the kernel that you can't load them as a module, they are clearly derived works just by virtue of being very intimate - and because the GPL expressly mentions linking).
So being a module is not a sign of not being a derived work. It's just one sign that _maybe_ it might have other arguments for why it isn't derived.
Linus
----8<----The real issue is if anyone who can claim copyright to Linux kernel will sue Canonical. I don't think so.
It's whether someone who distributes the kernel module and the kernel together is obliged to distribute the module under GPL terms as described in the "part of a whole" paragraph:
« If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it. »
(It seems pretty clear to me that the answer is "yes".)
Is it possible to glue a kernel module shim to something else under License A, glued to another shim under License B, etc until you have license compatibility with ZFS?
At the very least, this would seem to violate the very spirit of what the GPL intends.
Eben Moglen from Software Freedom Law Center thinks its OK as long as the source is distributed and no proprietary changes is made.
https://www.softwarefreedom.org/resources/2016/linux-kernel-...
>In this specific sense, then, the conduct which falls outside the words of GPLv2 falls within the "equity of the license," or its "spirit." As all Western legal systems have known since Aristotle, literal interpretation of any legal material will sometimes produce unintended unjust results, which can and should be corrected by the invocation of "equity." This present issue is evidently an example in which the tension between literal and equitable interpretation is raised, and it is the consensus of the kernel copyright holders' intention which determines which mode of interpretation is to be employed.
Linus is one central figure among kernel copyright holders. When he has stated his interpretation and it seems that most other maintainers are not objecting either, kernel copyright holders' intention seems to be closer to the "equity of the license".
Clearly ZFS is not a derivative work of Linux and Linux is not a derivative work of ZFS. Only the glue code is in violation (because it derives from both).
To determine whether there is glue here that violates the GPL would require looking at the particulars. Said glue might be licensed under the GPL, and might export the sorts of interfaces that Illumos does, thus allowing the ZFS module to link without reference to either the glue or the kernel! Then how could such glue be said to violate the GPL? It couldn't! Nor could ZFS be said to be violating the GPL in that case.
I don't think you can avoid an escape hatch for the GPL in the kernel without disallowing dynamic loading or modifying the GPL to have harsher terms. But recall that the GPL, and/or much software distributed under the GPL, is designed to have escape hatches: so you can run proprietary software using standard interfaces, for example (i.e., user-land code is not required to be GPLed just because it runs on a GPLed kernel, though a kernel could require it, it's just that _Linux_ doesn't).
It seems here you can't have your cake and eat it too.
I'm not sure where the legal line is going to get drawn, but clearly it's still fuzzy.
We each write something that passes as 'English' the language. Yet each of our brains has a different implementation of that interface. We do not copy each other even though we happen to be able to interact.
I'd like to see more legal argument, with citation to applicable precedent, on this conclusion. It necessarily, for functional reasons, makes some use of copyrighted kernel interfaces, but whether this use of functional elements makes it a legal derivative work is a question I've seen many people present cobflicting conclusions on, but never with much argument supporting the conclusion.
This then requires looking for a way in which the GPL prevents creating such adaptor modules. I don't see how it does. Perhaps it could if modified, but as it is I don't think it does. IANAL though.
In the kernel, functions which might be used by any modules are exported using EXPORT_SYMBOL. But if a module wants to use functions exported via EXPORT_SYMBOL_GPL, then the module has to use GPL.
This was brought up in a (very long) ML thread on LKML[1].
[1]: https://lists.linuxfoundation.org/pipermail/ksummit-discuss/...
If you read the GPL, it is clearly designed to prevent EXACTLY this thing. I'm sure as it was being written, Microsoft was envisioned as the evil entity that would steal free software and mix it with closed proprietary code. The license is clearly and deliberately designed to prevent this.
The "other side" may not see it this way. But the "other side" must abide by all involved licenses, including the GPL. Their own license, and the GPL license. The GPL is clear that if you cannot abide by all licenses, then don't do this.