Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
Intel Bug Graphics Open Source Upgrades Linux

Lots of Changes for Intel Graphics Coming in Linux 3.9 102

Posted by Unknown Lamer from the crush-all-insects dept.
With the Linux 3.8 merge over, the Intel Linux graphics developers are looking toward 3.9. From a weblog entry by one of them: "Let's first look at bit at the drm core changes: The headline item this time around is the reworked kernel modeset locking. Finally the kernel doesn't stall for a few frames while probing outputs in the background! ... For general robustness of our GEM implementation we've clarified the various gpu reset state transitions. This should prevent applications from crashing while a gpu reset is going on due to the kernel leaking that transitory state to userspace. Ville Syrjälä also started to fix up our handling of pageflips across gpu hangs so that compositors no longer get stuck after a reset. Unfortunately not all of his patches made it into 3.9. Somewhat related is Mika Kuoppala's work to fix bugs across the seqnqo wrap-around. And to make sure that those bugs won't pop up again he also added some testing infrastructure. " The thing I am most looking forward to is the gen4 relocation regression finally being fixed. No more GPU hangs when under heavy I/O load (the bane of my existence for a while now). The bug report is a good read if you think hunting for a tricky bug is fun.
This discussion has been archived. No new comments can be posted.

Lots of Changes for Intel Graphics Coming in Linux 3.9 More

Lots of Changes for Intel Graphics Coming in Linux 3.9

Comments Filter:

  • 3.8 has been released (Score:5, Informative)

    by Lennie ( 16154 ) on Monday February 18, 2013 @08:25PM (#42940561)

    The merge for 3.8 isn't just over, it has already been released.

  • If intel went into discrete graphics (Score:4, Interesting)

    by Duncan J Murray ( 1678632 ) on Monday February 18, 2013 @08:32PM (#42940597) Homepage

    I think they'd take over the linux graphics-card market. Maybe not much of a market now, but potentially could be big enough to justify doing it.

    • Not unless they dramatically increase performance. Many servers of course don't need much if any graphics, but a workstation or gaming box (think Steam for Linux) want more power than their current technologies can deliver.

      • Many servers of course don't need much if any graphics,

        People do scientific calculations on servers with lots of gpus.

        • Erm that would involve using a lot of graphics horse power.....which Intel doesn't have.
          That was the GP's entire point.

          • Intel has it. They're just not selling it to you. They have to figure out how to prevent us from running webserver VMs on this hardware before they release it. Unfortunately for them it's a lost cause. The people making these decisions really don't understand the mechanics of the situation, or how clever software can extract the utility of a GPU and deliver it to a cpu. Intel is now run by business geeks who really don't understand the tech. From here the end is clear.

        • Re: (Score:2)

          by dbIII ( 701233 )
          Only if it's something that doesn't require more memory than you can conveniently shovel into the GPU. It's the same reason why I couldn't use a cluster of playstations.

        • Re: (Score:2)

          by ssam ( 2723487 )

          intel have a very good contender for that specific market. the xeon phi.

          • xeon phi, the bastard child of larrabee? lol....

            Its too specific a piece of hardware, the code isnt portable to ANYTHING. stick to gpgpu with OpenCL....

            • Re: (Score:2)

              by ssam ( 2723487 )

              really? as I understand it can run openCL or openMP or MPI or a bunch of other standards.

        • Many servers of course don't need much if any graphics,

          People do scientific calculations on servers with lots of gpus.

          For Intel, wouldn't that be one of the case uses for the Itanium?

    • Re:If intel went into discrete graphics (Score:5, Insightful)

      by drinkypoo ( 153816 ) <drink@hyperlogos.org> on Monday February 18, 2013 @08:39PM (#42940649) Homepage Journal

      First, they would have to demonstrate more graphics aptitude than they have to date... Not all of us will accept inferior performance to get an untainted kernel.

    • Re: (Score:3)

      by Osgeld ( 1900440 )

      they used to be, and at the time you would pay as much for an intel video card as a older model nvidia, which would still stomp the shit out of it

    • Re: (Score:2)

      by smash ( 1351 )
      Why would intel go to "discrete" graphics when they could just build the same tech into their CPU and take advantage of local cache, not need to go over the PCI bus, etc?

      • Re: (Score:2)

        by Osgeld ( 1900440 )

        what century are you living in?

        • Re: (Score:2)

          by smash ( 1351 )
          What century are YOU living in? Integrated GPUs have come a long way and are improving faster than software requirements are advancing.

          • Re: (Score:2)

            by Osgeld ( 1900440 )

            the century where video cards are not on the PCI bus

            • PCI-express is still slower than it would be if video were integrated into the CPU so his point is valid even if his idea is likely unworkable.

    • Getting into discrete graphics they way you're thinking is a waste of money. Graphics cards are going the way of the dodo bird. The current trend is moving everything is moving onto the CPU. Onboard graphics has already cannibalized the low end graphics market, and is starting to cannibalize the mid-range. It's only a matter of time before the onboard tech catches up to the discrete tech.

      For the HPC market, Intel already has their MIC processors, latest being the Xeon Phi [wikipedia.org]. The only market they're not c

    • Re: (Score:2)

      by tlhIngan ( 30335 )

      I think they'd take over the linux graphics-card market. Maybe not much of a market now, but potentially could be big enough to justify doing it.

      Why? To gain what, an extra % of marketshare?

      Intel's already the largest GPU vendor out there by quantity. With very little exception, if you have an Intel CPU, you probably have Intel graphics coming along for the ride - companies like ATi^H^H^HAMD and nVidia are going after the leftovers.

      Intel's gotten there by being good enough to most people - their GPUs are ad

      • companies like ATi^H^H^HAMD and nVidia are going after the leftovers

        Much the same way that BMW, Mercedes, Audi, Porsche, etc are "after the leftovers" - i.e. they're only "leftovers" if you're somewhat confused when you consider the bigger picture...

  • GPU reset, Windows users should be so lucky (Score:4, Interesting)

    by Trax3001BBS ( 2368736 ) on Monday February 18, 2013 @09:10PM (#42940849) Homepage Journal

    Since Vista, the Windows OS deals with this by chatting with the graphics,
    if the graphics doesn't reply in a reasonable amount of time the OS will reset the program.

    Playing an intensive game like Battle Field 3 this tends to happen a lot. The video
    card is too busy and doesn't have the time to chat; so the game just goes away.
    Your either at your desktop like nothing is wrong other than not playing a game anymore,
    or staring at the last graphics frame shown with a hard reset in your future.

    VIDEO_TDR_FAILURE is what one error calls itself and fixed by disabling the "Timeout Detection and Recovery (TDR)."
    Key added here: HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\GraphicsDrivers then disabled.

    I have a write up about this in my journal but it's not worth reading, it's wordy and doesn't flow a bit.
    A better write up is here http://mikemstech.blogspot.com/2011/12/troubleshooting-0x116-videotdrerror.html [blogspot.com]
    a site I found after running a debug on the one (single) .DMP file that managed to be written before the OS
    became unresponsive.

    • Comment removed based on user account deletion

    • You have a problem (Score:5, Insightful)

      by Sycraft-fu ( 314770 ) on Monday February 18, 2013 @09:29PM (#42940941)

      You need to stop blaming Windows and/or the GPU vendor and troubleshoot your system. My GPU has reset precisely zero times when playing BF3, over a total of about 107 hours. I have seen GPU resets on my system on rare occasion, generally with broken software, but never in BF3 and I have more than a bit of testing with it.

      You have something wrong on your system, you should figure out what.

      • You need to stop blaming Windows and/or the GPU vendor and troubleshoot your system. My GPU has reset precisely zero times when playing BF3, over a total of about 107 hours. I have seen GPU resets on my system on rare occasion, generally with broken software, but never in BF3 and I have more than a bit of testing with it.

        You have something wrong on your system, you should figure out what.

        I'm not saying there's nothing is wrong with my system. I can say we have
        different systems I built mine not sure yours, but they are different.

        This GPU reset I refer to is intentional and part of a program called a "Windows user experience"
        "Windows User Experience Guidelines" http://www.microsoft.com/en-us/download/details.aspx?id=4249 [microsoft.com] (PDF)

        The GPU reset is explained here in more detail."Timeout Detection and Recovery of GPUs through WDDM"
        http://msdn.microsoft.com/en-us/library/windows/hardware/gg487368.asp [microsoft.com]

        • Re: (Score:1)

          by Anonymous Coward

          I'm not saying there's nothing is wrong with my system.

          But, there IS something wrong with your system. You got a shit video card. Happened to me here in China. Can't recall the exact Nvidia model, but the card frequently reset playing Team Fortress 2. I found under clocking the GPU by 50 Mhz increased the stability, but it wasn't till I bought a Nvidia GT440 for 300 yuan that the problem went away completely. Now, TF2 and a host of other games like Batman: Arkham City, GTA IV, Dead Space 2, Bioshock 2, and Bionic Commando play perfectly. I haven't had a s

          • I'm not saying there's nothing is wrong with my system.

            But, there IS something wrong with your system. You got a shit video card. Happened to me here in China. Can't recall the exact Nvidia model,

            EVGA GTX-570, can't believe nobody has had this happen or heard of it, and yes something could very well be wrong with my system (OS).
            The hardware itself is top notch.

            Between that registry entry and reinstalling the sound drivers once in a while I don't have problems any more.
            Can't have as much BF3 play time as I do and problems. not going to post the time played as it's excessive.

    • Re: (Score:2)

      by Z34107 ( 925136 )

      I haven't seen an error like that since the Longhorn beta. I don't doubt that you ran into that problem, but "it happens a lot" is just wrong.

    • And people say Windows is easier than Linux. :P

    • Try running some benchmark/testing software on your RAM. It sounds like you've some bad memory cells. And please be clipped to the case with an ESD wrist strap if/when you go poking around inside your case. I had a clients machine that would crash instantly if you waved your hand over the memory chips (dinosaur days). The ESD damage just accumulated enough that the system became flaky. I replaced the chips and the problem went away.

  • Not playing by the rules and standards dictated (Score:4, Informative)

    by Trax3001BBS ( 2368736 ) on Monday February 18, 2013 @09:12PM (#42940853) Homepage Journal

    GPU resets:

    "Other ideas for troubleshooting:"

    {list not quoted}

    "I wouldn't pin this problem on Microsoft. Ultimately, this crash is due to game/software developers
    and graphics card manufacturers (such as ATI/AMD and NVidia) developing buggy devices and software
    and not playing by the rules and standards dictated for a specific platform like Windows. There are
    many cases of similar events happening on UNIX/Linux systems, so this problem is not specifically isolated to Windows."

    The above is a quote from http://mikemstech.blogspot.com/2011/12/troubleshooting-0x116-videotdrerror.html [blogspot.com]
    a site I have nothing to do with; just a google result that helped me out one time.

  • Anyways...
    For years and years on end I used to study linux kernel code and hack it a few times.

    Not so much, anymore, these days.

    GEM's, Syrjälä's, seqnqo's --- by golly, I just hope you kids know what yer doing...

  • Wow, finally fixed? (Score:3)

    by ZorinLynx ( 31751 ) on Monday February 18, 2013 @09:51PM (#42941039) Homepage

    >the kernel doesn't stall for a few frames while probing outputs in the background!

    Is this why my Intel based XBMC box skips frames occasionally when watching videos? I was able to fix it by downgrading to a specific kernel version that did not have the problem.

    I've been unable to upgrade this machine ever since, and am still running on that ancient install and kernel for this reason. It'd be nice if it was finally fixed!

  • Unexploded acronyms (Score:4)

    by jabberw0k ( 62554 ) on Monday February 18, 2013 @10:05PM (#42941123) Homepage Journal
    Presumably the article refers to Direct Rendering Management [wikipedia.org] and not Digital Rights Management. Any clue what GEM they are talking about? Unless it's a revival of Digital Research's GEM [wikipedia.org] -- the alternate windowing platform that Ventura Publisher used to use. Article authors: Please expand acronyms with the first usage.

    • Re: (Score:1)

      by Anonymous Coward

      http://en.wikipedia.org/wiki/Graphics_Execution_Manager

    • Re: (Score:2)

      by geschild ( 43455 )

      I had the exact same association with GEM, but the link you provided really made me relive moments from over 25 odd years ago in college where I first saw that beast called GEM. Seeing your #id, I guess our lawns are in the same street.

  • Too bad the AMD sacked the vast majority of their software people with any sort of Linux skills/background at the end of October. They could have been relevant.

Slashdot Top Deals

Whoever dies with the most toys wins.

Close