Slashdot Mirror


Intel Starts Publishing Open-Source Linux Driver Code For Discrete GPUs (phoronix.com)

fstack writes: Intel is still a year out from releasing their first discrete graphics processors, but the company has begun publishing their open-source Linux GPU driver code. This week they began by publishing patches on top of their existing Intel Linux driver for supporting device local memory for dedicated video memory as part of their restructuring effort to support discrete graphics cards. Intel later confirmed this is the start of their open-source driver support for discrete graphics solutions. They have also begun working on Linux driver support for Adaptive-Sync and better reset recovery.

43 comments

  1. As long as the GPUs ... by CaptainDork · · Score: 1

    ... sign a non-disclosure agreement.

    --
    It little behooves the best of us to comment on the rest of us.
    1. Re: As long as the GPUs ... by CaptainDork · · Score: 1

      Then find another job, OK?

      --
      It little behooves the best of us to comment on the rest of us.
    2. Re: As long as the GPUs ... by Anonymous Coward · · Score: 0

      Then get out of my country and go to India, OK?

    3. Re: As long as the GPUs ... by CaptainDork · · Score: 1

      I'm no where near your country.

      I'm in Texas.

      --
      It little behooves the best of us to comment on the rest of us.
  2. Access your memory anywhere - with Intel inside! by Anonymous Coward · · Score: 0

    Why speculate? Just plug in the IME backdo.. integrated management system, you don't even need to turn it on - and boom - the entire internet has access to your bits and bytes. It's that simple! Now OSS flavored, slightly!

  3. I always thought it would be interesting by Solandri · · Score: 2

    and useful if a discrete GPU could begin to use system RAM as second-tier VRAM once the VRAM on board the GPU was exhausted. That would prevent the issue where if you run out of VRAM, the game starts to stutter as the game dumps textures from VRAM and is forced to read new textures in off disk. If those extra textures could be held in system RAM instead, the stutter when it was transferred to the GPU would be considerably smaller than having to read it off disk.

    Nvidia and AMD would never do this because it would cannibalize their sales of GPUs with more VRAM. Right now if your GPU doesn't have enough VRAM to run a game, your only choices are to reduce texture quality, or buy a new GPU. Intel only did it because they built GPUs without any VRAM, or with just 32-64 MB of eDRAM.

    The need has decreased as SSDs have supplanted HDDs. And some games appear to be doing this manually - caching all textures in system RAM so they don't need to be re-read from disk. But system RAM as second-tier VRAM would be faster and a more universal solution.

    1. Re:I always thought it would be interesting by Fly+Swatter · · Score: 4, Insightful

      They already use ram for additional storage, but the problem still exists because the speed of data transfer across the bus between the CPU and the GPU are still the limiting factor. Also the VRAM is tuned purely for the GPU, and is significantly more efficient even if you could stick a standard dimm slot onto the graphics card right next to the GPU.

    2. Re:I always thought it would be interesting by godrik · · Score: 1

      NVidia's GPUs already do that (I assume AMDs too).
      It is called unified memory in CUDA. It's been out there for years now.

    3. Re:I always thought it would be interesting by Anonymous Coward · · Score: 0

      if a discrete GPU could begin to use system RAM as second-tier VRAM once the VRAM on board the GPU was exhausted.

      That's been a feature since the introduction of AGP. It doesn't work nearly as well as you think it does.

    4. Re:I always thought it would be interesting by AHuxley · · Score: 1

      That memory was too slow for advanced new computer games.
      Some software likes lots of RAM when the math is well understood and the software can make use of the GPU and RAM.
      It depends on the task, math, skill and time put into the OS, GPU, of the software code.

      --
      Domestic spying is now "Benign Information Gathering"
    5. Re:I always thought it would be interesting by Anonymous Coward · · Score: 0

      Unified Virtual Memory (UVM) is only really achievable via NVLink. I think AMD can only achieve the same effect on systems which feature a kernel fusion driver (kfd) which pretty much means just Linux. There is no working kfd in Windows.

    6. Re:I always thought it would be interesting by Immerman · · Score: 1

      > even if you could stick a standard dimm slot onto the graphics card
      Now there's an idea. Sure, it'd be less efficient than VRAM, but it would be much more efficient than talking to main memory across the bus. And who doesn't have a few old DIMMS lying around, displaced by other upgrades?

      Of course the cost of a DIMM slot or two, and quite possibly a second memory controller to talk to the very different memory might very well outweigh the benefits

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    7. Re:I always thought it would be interesting by TeknoHog · · Score: 1

      This. See Intel740 for a GPU that practically ditched all onboard memory for AGP and failed miserably.

      --
      Escher was the first MC and Giger invented the HR department.
    8. Re: I always thought it would be interesting by Anonymous Coward · · Score: 0

      There are DDR4 GTX 1030 cards. They are substantially worse -- performance like half.

    9. Re: I always thought it would be interesting by Immerman · · Score: 1

      Absolutely. But if you've got a video card with high-speed RAM, *plus* lower speed add-on ram as an additional cache against having to transfer data over the bus, then that would almost certainly be an improvement over the high-speed ram alone.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    10. Re: I always thought it would be interesting by Anonymous Coward · · Score: 0

      You already have that. It's called THE REST OF THE FUCKING RAM.

      The RAM that's in those DIMMs on your motherboard.

    11. Re: I always thought it would be interesting by Immerman · · Score: 1

      What part of "over the bus" did you not understand? Memory buses are typically orders of magnitude faster than anything else outside the CPU. Typically in the range of 25-60GB/s. By contrast even a PCIe 3.0 lane has a bandwidth of just under 1GB/s. Even a 16x dedicated channel is going to have less than a third of the bandwidth of a fast memory controller, and FAR worse latency.

      That's the same reason your CPU has multiple layers of increasingly larger and slower cache sitting between it and RAM - even RAM is abysmally slow, especially when it comes to latency.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
  4. Excellent. by Gravis+Zero · · Score: 3, Insightful

    As much as I dislike Intel for their usual business practices, it's a good thing that they are bringing more open source hardware to the market. If nothing else, this will put additional preasure on other companies *cough*nvidia*cough* to be more open about their own hardware.

    I've always found it strange that some companies release hardware with almost no documentation and half-assed drivers because it's basically kneecapping your own product.

    --
    Anons need not reply. Questions end with a question mark.
    1. Re:Excellent. by Anonymous Coward · · Score: 0

      While the GPL zealots hate nvidia, they do release drivers for linux and freebsd. AMD does not support freebsd. Intel has a mixed track record with freebsd.

      It would be better if all three released open source drivers under a liberal license like mit, bsd, etc so that all systems could use it. That said, i support companies that support my platforms.

    2. Re:Excellent. by Anonymous Coward · · Score: 1

      Misleading. NVIDIA releases closed-source tarballs for BSD. AMD does not.

      AMD releases specifications and documentation and has great open source drivers. NVIDIA does not.

      Most of the work for both AMD and NVIDIA open source drivers are being done on Linux.

      NVIDIA doesn't support any platform except their own. AMD as well but at least they release the specifications to allow you to support your own platform.

      Intel and AMD follow a similar path. NVIDIA does not.

      Unless you've purchased a workstation graphics adapter for doing 3D CAD/CG/etc, you're a 'gamer' or 'amateur' and are whinging on sound bites gleaned from the media. Enjoy your nvidia experience. When they stop supporting 'your platforms' go buy another I guess.

    3. Re:Excellent. by Anonymous Coward · · Score: 0

      GPL zealots and FreeBSD do not belong to each others, dumb

  5. Comment by AdesanmiAdedotun · · Score: 0

    Linux is taking an edge way compared to other OS. It's high time Windows start to learn from these guys. WhatsApp Status BT notification for smartwatch Sci hub o2tvseries 192.168 ll

  6. Open Source NSA Backdoor? by Anonymous Coward · · Score: 0

    Go FUCKING HANG Intel. You betrayed freedom so fundamentally it no longer exists, thanks you Stazi cockguzzling fahhots! Mom is sooo proud!

  7. Really? Or did the just move it to a closed blob? by Anonymous Coward · · Score: 0

    We're constantly being deceived by these shitty companies. AMD (graphics are still dependent on proprietary components), Atheros (old wifi chips are good, the new ones like all newer wifi chips suck from a support perspective), and others have moved a lot of the code from drivers to proprietary firmwares over the years depriving us of having real genuine control over the hardware in our systems and this has prevented developers from fixing bugs, adding features, and just generally maintaining proper support for the hardware. Wake me when someone has confirmed that this isn't just another public relations stunt. Given Intel's GPU is supposed to be tech from AMD and AMD hasn't released a full set of code I'll be shocked if Intel has.

  8. Intel has a history of Linux support by drinkypoo · · Score: 1

    For example, superior support for power-saving as compared to AMD. AMD never bothered to properly support power-saving on e.g. my Athlon Mobile L110.

    --
    "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    1. Re:Intel has a history of Linux support by Anonymous Coward · · Score: 0

      That's bullshit, either OEM didn't implement proper BIOS settings or user/idiot error. Either way neither one is AMD's fault - AND you trying to segue into that from "intel has a history of linux support" = laughable trolling, you're a putz.

    2. Re:Intel has a history of Linux support by Anonymous Coward · · Score: 0

      You might want to read
      https://bugzilla.kernel.org/show_bug.cgi?id=109051

    3. Re:Intel has a history of Linux support by drinkypoo · · Score: 1

      That's bullshit, either OEM didn't implement proper BIOS settings or user/idiot error.

      You talk a lot of shit for a coward.

      Either way neither one is AMD's fault

      Intel commits code to support their processors and chipsets. AMD didn't bother to do that for this chip, so it had to be guessed and bodged from other chips.

      AND you trying to segue into that from "intel has a history of linux support" = laughable trolling, you're a putz.

      I use AMD processors pretty much exclusively. I am typing this from a system with FX-8350 right now. If facts frighten you, perhaps you should fuck right off.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    4. Re:Intel has a history of Linux support by Anonymous Coward · · Score: 0

      So idiot/user error then. As called. You have no idea how to implement power saving features regardless of OS and that's AMD's problem somehow. Whatever, putz. They do actually commit code to Linux though, you know nothing.

  9. WOW by Anonymous Coward · · Score: 0

    I could not give less of a shit

  10. Not open if the firmware is locked down. by Anonymous Coward · · Score: 0

    Same as AMD and Nvidia.

    And even more dangerous is the serials/GUIDs they can pass to software, possibly without you even having library level filtering of it.

    Each little bit makes it that much harder to stay anonymous, whether online, or gaming. Do you really trust Big Data/Big Companies to look out for YOU?

    Even in cyberpunk, while they might have known everything about you, they couldn't perfectly simulate how you would think or definitively fingerprint your online identity from normal user activity. Today however that is becoming increasingly true.

  11. A row of CPU's by AHuxley · · Score: 1

    A big long card with a few CPU's and sell it as a new way of thinking about a GPU.
    Existing CPU design trying to sell ray tracing as a new powerful GPU design.
    Can all todays GPU math be made extra fast by using a lot more CPU math?
    Fast CPU math will make an amazing GPU card for a set of ray tracing math.
    CPU math that computer games will have to understand and work to support as graphics.
    Just keep adding another CPU onto the GPU card until the rays work at 60 fps in 4K?
    All games crave adding that extra open source Intel ray tracing math...

    --
    Domestic spying is now "Benign Information Gathering"
    1. Re:A row of CPU's by Anonymous Coward · · Score: 0

      Intel already tried that. It was called Larrabee. It failed, and became Xeon Phi, which is also now pushing up daisies (Phi is replaced by Cascade Lake-AP):

      https://www.anandtech.com/show/3738/intel-kills-larrabee-gpu-will-not-bring-a-discrete-graphics-product-to-market

      https://www.nextplatform.com/2018/07/27/end-of-the-line-for-xeon-phi-its-all-xeon-from-here/

  12. Hope their drivers will be more stable than AMD's by ffkom · · Score: 1

    As much as I appreciate AMD's efforts to implement the "amdgpu" driver, the result is still so far away from being stable enough for serious 24/7 production use (rather than just gaming), that I really hope Intel will do better and provide an alternative for buyers.

    After all, the i915 has been very reliable for me in recent years.

  13. First discrete graphics processor by Anonymous Coward · · Score: 0

    What about Intel's i740 from 1998?

  14. WHEN THE TOUGH GET GOING (NVDA/AMD) THE WEAK by Anonymous Coward · · Score: 0

    (INTL) GET DESPERATE.

  15. Re:Really? Or did the just move it to a closed blo by Anonymous Coward · · Score: 0

    Intel's GPU is not "supposed to be tech from AMD". You're probably thinking of Kabylake-G, which is not related to this discussion.

    Intel's new dGPU is known as Intel Xe. The tech behind it is an iterative extension of their "Gen" graphics used in their iGPUs, which are in turn heavily modified from patents they licensed from nVidia years ago.

    Gen11 is the last "Gen" graphics iteration. It will appear in IceLake processors, starting in Q4 of this year. Gen12 was renamed to Intel Xe. Expect whatever is the successor to IceLake to have Xe for its iGPU. Intel's dGPUs will also be based on Xe.

  16. What About Spectre and Meltdown? by Anonymous Coward · · Score: 0

    Shouldn't those be the highest priority issues to be addressed?

    1. Re: What About Spectre and Meltdown? by Anonymous Coward · · Score: 0

      new features/products bring more money and a lot more new clients, bug fixes rarely do.

  17. In other words by TeknoHog · · Score: 2

    Some companies like to wait until product launch, but Intel isn't being too discrete about their plans.

    --
    Escher was the first MC and Giger invented the HR department.
  18. 3Dfx Voodoo by Anonymous Coward · · Score: 0

    I really didnâ(TM)t think those cards and chips would go anywhere. The target audience is gamer kids who donâ(TM)t have jobs. I was so wrong, amazing.