No announcement yet.

Wayland, Dumb Frame-Buffers & Embedded SoCs

  • Filter
  • Time
  • Show
Clear All
new posts

  • Wayland, Dumb Frame-Buffers & Embedded SoCs

    Phoronix: Wayland, Dumb Frame-Buffers & Embedded SoCs

    If you're not following the many Linux development mailing lists out there, the latest major discussion surrounding the Wayland Display Server that's spanned the Wayland, DRI/DRM, and Fbdev mailing lists has been about using Wayland on "dumb frame-buffers", KMS vs. fbdev, and DRM drivers on embedded SoCs...

  • #2
    Thanks for summarizing the situation, now let's hope this improves...


    Thanks for publishing this article, I hope it helps to make the situation gain some visibility and hackers look at possible solutions. I asked a bit with my limited skills about this, in hope to heat up the debate a bit and see if current hackers start to think about it in a public form again. Then I wrote a fastly written forum message here in hope to make it easier to gain some visibility from the Phoronix staff, it seems it worked

    The news are quite interesting, but probably some users did not understand the importance of this. This not only covers Wayland or X, but also it can help to improve the current binary blob mess in the embedded Linux platforms.

    Maybe some further clarification could be interesting, as I think those topics needs a divulgative introduction of the different topics coveted in the article. Michael and other Phoronix editors: What do you all think about it? I think people are underlooking KMS these days, it's a killer app not only useful for better console to X switching but even more.

    I hope someday we get to be able to use KMS to it's full potential, full OpenGL support in parity with other major platforms without being second citizens (this means supporting each new version early), open source graphical drivers get nearly as good as closed source ones and the binary blob mess stops from the embedded world. In definitive making Linux more true to the Free/Open Software philosophy and a lot stronger in terms of graphics. In Linus words...

    Originally posted by Torvalds, Linus (2007-03-19). The Torvalds Transcript: Why I 'Absolutely Love' GPL Version 2
    Me, I just don't care about proprietary software. It's not "evil" or "immoral," it just doesn't matter. I think that Open Source can do better, and I'm willing to put my money where my mouth is by working on Open Source, but it's not a crusade ? it's just a superior way of working together and generating code.
    It's superior because it's a lot more fun and because it makes cooperation much easier (no silly NDA's or artificial barriers to innovation like in a proprietary setting), and I think Open Source is the right thing to do the same way I believe science is better than alchemy. Like science, Open Source allows people to build on a solid base of previous knowledge, without some silly hiding.
    But I don't think you need to think that alchemy is "evil." It's just pointless because you can obviously never do as well in a closed environment as you can with open scientific methods.
    I'm just a Linux user, but I did see the graphical ecosystem of Linux needs a big shake up. Fortunately there's stuf like Wayland and KMS, but there's lots of improvements needed and also reunificate others. If I was able to understand that with all my limitations, I'm sure skilled people can see a better image of the situation and help improving it in a radical way.

    I see the grow of graphical memory managers is too excessive, the TTM vs GEM situation got a bit ridicule from the user POV in the past, but unfortunately the situation is a lot worser because there are lots others that are out of tree. I hope someday all those get replaced by an unified and powerful memory manager into the vanilla tree of Linux.

    Another interesting last reply:Laurent replied this...
    Originally posted by Laurent Pinchart
    We're also evaluating the possibility of providing a generic fbdev emulation
    layer on top of V4L2 without requiring any device-specific fbdev code. fbdev
    isn't maintained and hasn't really evolved for quite some time now.

    I'm not sure if this is good or bad in different means. I think Linux's Frame Buffer was a quite good idea in the past, but it got dirty over the years. KMS is the new king and it needs to be promoted to being the improved successor.

    Anyway... why Video for Linux growed so big? Why that subsystem can't be modularized and used in a really sane way? Why such total mess?

    We users would like to understand this situation a bit better, follow what is being done to solve it in a well explained way (divulgative efforts are very positive in the community, some of those users and someday be developers helping to improve things or they can start looking at those topics being already developers). I think that's why sites like Phoronix exists and I appreciate a lot the effort put into them.


    PS: As you see, most forum posts are cross-posted to different mailing lists (dri-devel, wayland-devel, linux-fbdev and some to linaro-dev and linux-media). This is due to the importance of the topic to different graphics-related projects and the involment of their developers in the task to solving the current total mess in different stuff.


    • #3

      I have some typos (it's usual in me, as I'm not native English), but I have one important there.

      I did mean mailing list messages, not forum ones. Sorry, I did write that very fastly (and unable to edit it later due to the one minute limitation, it should be at least 4).