Re:GPL Violations Related to Combining ZFS and Linux
by ajdlinux on Friday February 26, 2016 @08:04AM
(#11181134)
Of the 9 commits listed there, 2 are removing LGPL headers, 2 are adding LGPL headers to scripts which aren't linked into the kernel, 1 is relicensing a header file so it can be used outside the kernel, 1 is changing from linking against an external GPL3 library to an LGPL2 library without changing the licence of the tool that's in the kernel tree, 1 is an author noting that he's bringing in some external LGPL code which he wrote and is relicensing, 1 is a commit removing a driver and noting that an LGPLed userspace replacement exists, and 1 is noting an external LGPL userspace tool in the MAINTAINERS file.
So of all of those, literally 1 is changing the licence to LGPL.
A better search is https://github.com/torvalds/linux/search?utf8=%E2%9C%93&q=%2..., which reveals 147 results in the code.
This is still tiny, given that the kernel is a ~20 million LOC codebase, so yes, I would call LGPL code in the kernel an unusual thing.
Re:GPL Violations Related to Combining ZFS and Linux
by rcthompson on Thursday February 25, 2016 @07:10PM
(#11178925)
Hmm. I suppose if you could show that the Linux kernel source files required to build zfs.ko were all LGPL, then distributing it as a binary module might actually be legal.
I'm not well enough traversed in the kernel source to find a direct reference but you see commits like this every now and then:
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux....
Having parts of the kernel LGPLed doesn't seem like an unusual or odd thing, from what I can tell:
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux....