News for hackers. Stuff that matters.
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by lelandbatey on Monday September 16, 2013 @02:08PM (#6394737)
I'm confused. I've seen this a bunch of times, and I don't understand why Canonical submitted a patch to a video driver to "support" their Xmir window server. Why does a video driver need to have code in it to support a window server? Isn't it supposed to be the other way around? Isn't the window server supposed to work on top of the video driver (and all the other hardware drivers)? What's going on?
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by tinco on Monday September 16, 2013 @02:25PM (#6394849)
On Linux it is actually the window server that provides the interface between applications and the video driver.

Linux does not have a standard graphics driver interface like Windows does, at the moment is sort of defined by X, and since Wayland and Mir aim to replace X, the interface will be replaced too.

This means that every driver needs to interface with each of the window managers.

I would agree that is a silly idea, but such is the world when applications need very low level access to hardware to attain real time performance. You just get a tangly mess.

Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by jmillikin on Monday September 16, 2013 @02:46PM (#6395010)

  > On Linux it is actually the window server that provides
  > the interface between applications and the video driver.
This hasn't been true for a long, long time. Modern Linux applications directly call into the graphics driver, either through OpenGL or an intermediate library such as Clutter.

  > This means that every driver needs to interface with
  > each of the window managers.
This is also not true. Window managers are driver-independent, and you can easily write your own window manager without any knowledge of the underlying hardware. You might mean "display server" instead of "window manager", but there's only one display server used in Linux, and that is X11.

X11 was designed in an era before standardized graphics or input APIs, so it doesn't use OpenGL. As a result, it needs its own set of drivers for nearly everything (keyboards, mice, graphics, overlays). That's why Linux graphics drivers need to have special X11 support.

Wayland and Mir are an attempt to replace X11 with a much thinner layer, handling only the bare necessities of getting the driver and user application together so they can talk OpenGL together.

Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by tinco on Monday September 16, 2013 @07:31PM (#6396603)
You are right, I meant 'display server' instead of both 'window server' and 'window managers'.
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by lelandbatey on Monday September 16, 2013 @02:32PM (#6394909)
Ok, that makes more sense (or at least explains why). I'm puzzled that there hasn't ever been anything done to make this less of a tangly mess; it seems like it'd be a huge step towards making linux a better platform to iron something like this out. However, that's probably just wishful thinking, as I have no idea what it'd take technically or politically to make that happen.
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by tinco on Monday September 16, 2013 @02:36PM (#6394936)
There have actually been many tries, Wayland being the most recent of them and now Mir. But because the earlier ones relied too much on the old X code, they were not succesful, and Wayland has not yet secured support from the graphics cards manufacturers (except Intel). Mir is promised to change all that, but we will have to see.
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by voyou on Monday September 16, 2013 @02:59PM (#6395085)
Yes, I don't understand this, either. Note that a) the Intel X driver does not, AFAICT, have any special code to support Wayland and b) the code Ubuntu want to add appears to be for XMir, which is an X server that runs on top of Mir, not for the Mir display server itself. An X server running on top of a different display server should not need specific support in the drivers of the X display server; that XMir apparently does makes me worried Mir is going to be an architectural clusterfuck.
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by gue5t on Monday September 16, 2013 @05:52PM (#6396155)
It actually does have code to support XWayland, the analogue to XMir: <http://cgit.freedesktop.org/xorg/driver/xf86-video-intel/log...

This code is an optimization, not a strict requirement, as you're right that in theory XWayland or XMir could be supported in a hardware-agnostic manner. For XWayland this is the (formerly "wlshm") xf86-video-wayland X driver: <http://cgit.freedesktop.org/xorg/driver/xf86-video-wayland/>, but it is much slower than the hardware-specific options (which also exist for the radeon and nouveau X drivers). I don't know if there's an equivalent hardware-agnostic X driver for Mir, but in no case would a production-quality windowing server have that as its only method of compatibility for an important hardware segment--both Wayland and Mir developers expect X-dependent applications to stick around for quite some time, so they must perform as well as possible.

For what it's worth, my hat is firmly in the Wayland camp; its design is simpler, its development more open, and its motivations more clear than those of Mir. Canonical does a bad job of software stewardship and I'd hate to see them in control of the dominant graphics solution for Linux.

Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by ihsw on Monday September 16, 2013 @02:25PM (#6394856)
Beyond that, it may be because Mir is GPLv3 licensed and Wayland is MIT licensed. Just a hunch.
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by tinco on Monday September 16, 2013 @02:37PM (#6394950)
Oh I didn't even know that. That's a great reason to prefer Wayland over Mir or Mir over Wayland. I'm a bit torn now..
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by ihsw on Monday September 16, 2013 @02:49PM (#6395021)
Indeed, also X.Org is X11 Licensed (MIT license variant).

Wayland and Mir driver support is pretty much a mystery, and Canonical has been trying to engage hardware designers (Nvidia, Intel, AMD) for support. This recent spate between Canonical and Intel may be indicative that the process isn't going well.

Ubuntu's next LTS release is 14.04 so it's 8 months away, and if there's any surefire way to piss Canonical off then Intel us definitely getting that job done. I'm not trying to defend Canonical but it's hard to ignore the reality of the situation.

I wouldn't be surprised if BSD licensing is on the table front-and-center during negotiations, considering the legal barriers surrounding GPLv3.

Also, it makes me wonder where this leaves Valve. I get the distinct impression that they've been getting cozy with Canonical and they might be getting nervous about the situation.

Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by tinco on Tuesday September 17, 2013 @11:34AM (#6399854)
Hmm, I would say the oppposite. The recent spate between Canonical and Intel might indicate that Canonical's engagement with competitors NVidia and AMD are going well. Canonical doesn't really need to cooperate with Intel, because Intel already makes excellent open source drivers for their graphics cards.

I think Valve is the reason that Canonical can work together with NVidia and AMD, and that they're probably not nervous about this at all.

Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by tga_d on Monday September 16, 2013 @03:11PM (#6395159)
Not really. I much prefer the GPL, since I think copyleft is (at least partly) what has made Linux the massive success it is, and I think a copyleft licence for a display server would similarly be better. But in this situation, I have to support Wayland over Mir. The reason largely being that Wayland vs. Mir actually gives credence to the argument that Linux can't succeed on the desktop because of fragmentation. Normally in response to that argument, I'd say it's not that big of a deal, and that things like different package managers and desktop environments keep a healthy level of competition, while still having a great amount of cooperation. But when it comes to display servers, we're talking about something that requires driver support, which is already one of (if not the) top problems on GNU/Linux. Forcing the small teams that deal with Linux graphics driver development to choose between using their already limited time to support the general standard or what the most popular distro uses is a horrible decision.
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by spacelizard on Monday September 16, 2013 @03:05PM (#6395128)
Not really. GPLv3 will prevent hardware manufacturers from locking down their devices if they intend to ship Ubuntu Mobile. This is a good thing for users of those devices. I can't defend Canonical's support of proprietary drivers, especially on these devices that are known to be used for such pervasive spying, but at least a non-hardware-locked device allows for workarounds (i.e. deployment of your own free drivers).
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by wmf on Monday September 16, 2013 @03:48PM (#6395454)
GPLv3 will prevent hardware manufacturers from locking down their devices if they intend to ship Ubuntu Mobile.

Actually the opposite; Canonical will sell phone vendors a non-GPL version of Mir so that they can lock down their phones.

Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by ldng on Monday September 16, 2013 @04:23PM (#6395646)
Hence the copyright assignment required for any contribution to Canonical's projects, to ease the double licensing.

But for now, it's only speculation. Time will tell.

Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by mhall119 on Tuesday September 17, 2013 @12:56AM (#6397653)
It's not copyright assignment, it's just a broader license grant than what the GPL allows.
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by tga_d on Monday September 16, 2013 @03:16PM (#6395191)
I'm sorry, but must be missing something. How does a GPLv3 display server prevent locking down the hardware? Afaik, the anti-TiVoization stuff only applies to the software that's GPLv3, not the stack it's running on.
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by dkuntz2 on Monday September 16, 2013 @07:16PM (#6396531)
Some argue that if it makes calls to a GPL application, it's a derivative work, which means it must also be GPL.

This means that if one part of the stack is GPL, everything higher must also be GPL.

Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by eltondegeneres on Monday September 16, 2013 @09:54PM (#6397141)
Can you point me towards some people arguing this? It's patently false. If it were true, every single application on your typical GNU/Linux box would also be GPL.
Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by achat on Monday September 16, 2013 @02:23PM (#6394837)
While developing some application, did you ever feel that your SDK is lacking some feature which would make your app development much easier?

So think the video driver as library and the windowing system as an application using video driver library.

Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by lelandbatey on Monday September 16, 2013 @02:30PM (#6394888)
Yep. However, if you're trying to replicate something that's already built on top of an existing piece of software using the existing API's, I'd assume you wouldn't have to patch the platform just to build on top of it.

However, as tinco explained[0], it seems that there is no "standard graphics api" for linux, and the "link" between window servers and graphics drivers is hard coded. So you do in fact have to patch the video driver to provide support for a new window server (on Linux at least).

Re:Intel rejection of Ubuntu’s Mir patch forces Canonical to... by route66 on Monday September 16, 2013 @02:30PM (#6394886)
It's outlined here: http://www.x.org/wiki/guide/concepts/#xisclientserver . In a nutshell, X "serves" the application to the (in- and output) devices.