"Linux has a large amount of device drivers for hardware not supported on FreeBSD, especially USB devices. Not rarely, such drivers have been written based on information derived by protocol sniffing, reverse engineering and the like. This makes the code highly undocumented, and renders the porting effort extremely error prone. To help with this task, I decided to start working on an emulation layer that would let us recompile the linux source code on FreeBSD, and provide a sufficiently complete emulation of the kernel APIs so that device drivers (or at least certain classes) could be used without modifications to their source code."
To view parent comment, click here.
To read all comments associated with this story, please click here.
Member since:2006-01-21
Butters is right. This "compile by yourself is legal" clause is one (if not the main) reason why the attempt to block all binary kernel modules as of 2008/01/01 suggested by GKH recently was turned down[1].
With respect to the binary video drivers from NVIDIA and ATI the two crucial points are
- That distributions can't distribute the already linked binary kernel modules (like for example in the incident with the Kororaa Live CD[2])
- Whether the wrapper code for the binary blob itself can be considered a derivative work of the kernel, or just a code "using the structures provided by the kernel"
I have little experience with the latter issue (although it seems to be a proverbial grey area that is here to remain), but the first is almost clearly in violation of the GPL and should therefore be avoided.
Hope that clears things up
[1]http://article.gmane.org/gmane.linux.kernel/475890
[2]http://en.wikipedia.org/wiki/Kororaa#Kororaa_XGL_Live_CD_and_the_GP...
Member since:2005-07-13
According to butters, the license of the kernel module doesn't matter, as long as you don't distribute the linked binary. And that's exactly what NVidia is doing: they're not distributing linked binary kernel modules.
Actually, they are. nVidia does make binary modules compiled against specific versions of Suse distribution kernels available for download, for instance.
Their legal department seems comfortable with the fact that their binary blob is not derivative of the kernel, since it is OS-agnostic and the exact same blob is used on Windows and *nix. I suspect their legal department is correct.
Member since:2006-09-21
Actually, AFAIK -- and someone correct me if I'm wrong -- the binary modules are NOT LINKED (even when "compiled against" or basically using the kernel source/headers from a specific distro). I think all drivers for *nix OSs are like that: they are essentially like *.o object files and are NOT LINKED until loaded into the kernel.
There is not a pre-established dependency in the *.ko or drivers of where to find its external functions. In other words, there is no nothing in the module themselves that state they strictly require a linux kernel. It's theoretically possible to load and link those modules against a different kernel. To "link" means to establish a dependency for symbol resolution. But in these cases, the drivers don't care where they find their external symbols as long as they're found.
On the other hand, compiling the modules into the kernel as part of the kernel boot image then distributing it, would I think violate the GPL.
The Verge pointed me to a blog post by Leap Motion - which reveals how their Kinect-like motion control works with Windows 8. "From the second you plug in your Leap Motion Controller, you'll be able to browse the web and interact with your computer just by moving your hands and fingers in the air. With Leap Motion technology and Windows, you can do everything that's possible with multi-touch inputs - without actually touching anything. This also means that existing applications in Windows 7 and 8 will respond to your natural hand and finger movements. Soon, we'll show you how Leap Motion will work with Mac OS X." Quite cool.
PocketNow interviews Marc Dillon, and there's an interesting note about why Jolla is keeping the display properties under wraps: "We're leaving some of those details out because we do understand that there are a lot of really big players in the market and they tend to take certain components in the market and dominate them. We created the ability to actually be able to run Sailfish on multiple hardware displays and be able to swap components, so this is part of the demand and supply planning phase. We are committed to this industrial design which is a 4.5-inch display, an 8 megapixel camera on the back and a front-facing camera at the front, and the exact specs of the display we'll provide when we're close to delivery." Something you rarely hear anything about.
"This is my NeXTstation Turbo Color booting the Mac System 7.1 Operating System using a hardware add-on called 'Daydream'. The Daydream has a Mac ROM and a few custom ICs in it and hooked up to the NeXTs DSP port. Turns your NeXT into a full fledged 68k powered Mac." So cool.
"The NetBSD Project is pleased to announce NetBSD 6.1, the first feature update of the NetBSD 6 release branch. It represents a selected subset of fixes deemed important for security or stability reasons, as well as new features and enhancements. Please note that all fixes in the prior security/bugfix updates (NetBSD 6.0.1 and 6.0.2) are also in 6.1."
"At the end of a week in which Electronic Arts confirmed it wasn't developing a thing for the Wii U, one of the software engineers in EA Sports' Canada studio, in a series of since-deleted tweets, disparaged the console as 'crap' and suggested Nintendo should give up on hardware altogether. 'The Wii U is crap. Less powerful than an Xbox 360. Poor online/store. Weird tablet', tweeted Bob Summerwill, listed as a senior software engineer at EA Canada, in a reply to a tweet posting a link about EA's no-Wii U news. 'Nintendo are walking dead at this point'." The Wii U is turning into the 21st century's Virtual Boy.
"In NixOS, the entire operating system - the kernel, applications, system packages, configuration files, and so on - is built by the Nix package manager from a description in a purely functional build language. The fact that it's purely functional essentially means that building a new configuration cannot overwrite previous configurations. Most of the other features follow from this." Interesting approach. A Linux distribution, sure, but with some very refreshing ideas about system configuration.
Appfour added, among other features, C/C++ support to its new version of AIDE. From Android-IDE, "Now you can write parts of your app or your whole app in C/C++ on your device. AIDE supports the Standard Android NDK toolchain (GCC 4.6 + Bionic, STL, ...). No changes are necessary if you want to build an app developed on a PC with Eclipse. C/C++ development is fully integrated: Build errors appear in the error list and files can easily be navigated to with Go to file. The editor supports C/C++ syntax highlighting."
Ars nails it: "The answer is that Google did announce what amounts to a fairly substantial Android update yesterday. They simply did it without adding to the update fragmentation problems that continue to plague the platform. By focusing on these changes and not the apparently-waiting-in-the-wings update to the core software, Google is showing us one of the ways in which it's trying to fix the update problem."
"It is good for programmers to understand what goes on inside a processor. The CPU is at the heart of our career. What goes on inside the CPU? How long does it take for one instruction to run? What does it mean when a new CPU has a 12-stage pipeline, or 18-stage pipeline, or even a 'deep' 31-stage pipeline? Programs generally treat the CPU as a black box. Instructions go into the box in order, instructions come out of the box in order, and some processing magic happens inside. As a programmer, it is useful to learn what happens inside the box. This is especially true if you will be working on tasks like program optimization. If you don't know what is going on inside the CPU, how can you optimize for it? This article is about what goes on inside the x86 processor's deep pipeline."
"It was the only moment I heard regret slip into Otellini's voice during the several hours of conversations I had with him. 'The lesson I took away from that was, while we like to speak with data around here, so many times in my career I've ended up making decisions with my gut, and I should have followed my gut,' he said. 'My gut told me to say yes.'" The world would've been a much different place - Apple would have been less dependant on Samsung for its chips, which probably would've meant less money for Samsung to develop its Galaxy business.
Member since:
2005-11-11
Uhm... no? I quote butters:
"Probably by using the (unfortunately not so well known) fact that NO free software license places restrictions on how you USE the code (as opposed to distributing the code). Rest assured that you are free to link code licensed under incompatible licenses to your heart's content as long as you don't distribute the linked binary."
According to butters, the license of the kernel module doesn't matter, as long as you don't distribute the linked binary. And that's exactly what NVidia is doing: they're not distributing linked binary kernel modules.