Ubuntu Working Towards A Rootless X Server

[nanonyme]

I don't think it would be feasible. KMS is integrated right into the kernel, ATI/NV's binary blobs, OTOH, try as hard to decouple from the OS because they came from Windows Drivers, that's why it would be unfeasible for them to provide KMS binary drivers: too much of work.

Also I doubt it's even necessary. They "just" need to make sure everything that requires hardware access is done in their proprietary kernel module and userspace uses ioctl's and such as an uprivileged user. Something similar to KMS but not actually using KMS API. Both nVidia and ATi likely have most if not all if it in place since they've had kernel memory management for a while, though they'd need to go through a round of scrutinous testing to find out whether there's some remainders in userspace that would break if X would suddenly stop being run as root.

[unix_epoch]

Third, even if they come up with some plan to seed Linux kernel with a KMS interface that would work with either binary blobs, do you think Linus the ego man will accept it into the kernel? No, over his dead corpse! So No, no and no.

My guess is Linus isn't the problem (he doesn't seem too opposed to binary blobs based on my reading), but rather the other kernel developers who either disagree with his interpretation of the GPL2 or oppose the binary drivers on moral or maintainability grounds.

Personally, I'd love to see a lightweight interface to KMS that allows the nvidia and fglrx drivers to get along with the framebuffer and provide a seamless boot sequence (great for HTPCs, kiosks, x86 embedded devices, etc.).

[deanjo]

My guess is Linus isn't the problem (he doesn't seem too opposed to binary blobs based on my reading), but rather the other kernel developers who either disagree with his interpretation of the GPL2 or oppose the binary drivers on moral or maintainability grounds.

Exactly, Linus isn't opposed to blobs, he doesn't love them but he is also rational and sees the need and if linux is to be successful a compromise has to be made.

[airlied]

Exactly, Linus isn't opposed to blobs, he doesn't love them but he is also rational and sees the need and if linux is to be successful a compromise has to be made.

Lols, you actually heard Linus say that? citation needed.

Linus doesn't approve or disprove of binary blobs, the kernel is under a license and Linus distributes it under that license, he respects the GPLv2 and what it allows. He doesn't give a crap about linux success via compromise. The fact that you could attribute that to him is pretty funny and is probably some sort of justification you give yourself for compromising your own system by running binary drivers.

All binary blobs are not equal. Its all about derived works, some binary blobs are obviously derived works of the Linux kernel and hence illegal to distribute, some are not legally derived works and thus are legal to distribute, as long as they aren't linked before distributions.

Lawyers/judges get to decide where the legal lines lie not Linus. Other developers have a habit of thinking they understand the GPLv2 better or want to make it more obvious what a derived work is (hence EXPORT_SYMBOL_GPL).

Dave.

[deanjo]

Lols, you actually heard Linus say that? citation needed.

Google "shades of gray" with Linus's name. You will find some references in the LKML referring to what I'm talking about, or search these forums where it has been quoted a few times here.

[deanjo]

is probably some sort of justification you give yourself for compromising your own system by running binary drivers.

The only justification I need for that is that the alternatives are slower, featureless and not practical with my use.

[airlied]

Google "shades of gray" with Linus's name. You will find some references in the LKML referring to what I'm talking about, or search these forums where it has been quoted a few times here.

"he is also rational and sees the need and if linux is to be successful a compromise has to be made."

this is not the same as shades of gray, this is your interpretation of what Linus has done.

All Linus states is that there are cases where under the GPLv2 people can link binary blobs to the Linux kernel legally. This isn't a compromise for success, its just the reality of the license he choose. I did mean to ask him if there was a license that stated that prohibited linking non-derived blobs with the kernel existed at the time would he have used it. But it isn't like he picked GPLv2 because it allowed this feature or as some sort of compromise.

Dave.

[d4ddi0]

We already got a Rootless X

"Rootless"
I just logged in to say:

I don't think that means what you think it means.

Everybody else pretty much covered it, including the fact that this misnomerization (see I can abuse the language too!) originated at canonical.
How about non-root? or non-privileged? I already have Rootless X running in cygwin on my Windoze box at work.

[deanjo]

"he is also rational and sees the need and if linux is to be successful a compromise has to be made."

this is not the same as shades of gray, this is your interpretation of what Linus has done.

All Linus states is that there are cases where under the GPLv2 people can link binary blobs to the Linux kernel legally. This isn't a compromise for success, its just the reality of the license he choose. I did mean to ask him if there was a license that stated that prohibited linking non-derived blobs with the kernel existed at the time would he have used it. But it isn't like he picked GPLv2 because it allowed this feature or as some sort of compromise.

Dave.

Just the mere fact that Linus cares more about technology then licenses shows that he is committed to giving the end user the best end experience should speak volumes of what he cares about.

What was the point again?

Was the point to alienate people by showing how we're less about the
technology than about licenses?
......

We should make decisions on TECHNICAL MERIT.

......

I happen to believe that there shouldn't be technical measures that keep
me from watching my DVD or listening to my music on whatever device I damn
well please. Fair use, man. But it should go the other way too: we should
not try to assert _our_ copyright rules on other peoples code that wasn't
derived from ours, or assert _our_ technical measures that keep people
from combining things their way.

If people take our code, they'd better behave according to our rules. But
we shouldn't have to behave according to the RIAA rules just because we
_listen_ to their music. Similarly, nobody should be forced to behave
according to our rules just because they _use_ our system.

There's a big difference between "copy" and "use". It's exatcly the same
issue whether it's music or code. You can't re-distribute other peoples
music (becuase it's _their_ copyright), but they shouldn't put limits on
how you personally _use_ it (because it's _your_ life).

Same goes for code. Copyright is about _distribution_, not about use. We
shouldn't limit how people use the code.

--- Linus Torvalds

[RahulSundaram]

Hi

False dichotomy. Caring about licenses does not mean we don't care about technology or vice versa. None of the free and open source licenses restrict usage in any way. A pure copyright license just doesn't restrict usage at all.

Linus position is not a compromise. It is just a simple acceptance of the reality of the license.