Announcement

Collapse
No announcement yet.

AMD Publishes Open-Source Linux HSA Kernel Driver

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • tomtomme
    replied
    Originally posted by asdfblah View Post
    http://www.phoronix.com/scan.php?pag...tem&px=MTY0MDY

    How does this benefit other architectures, and OpenCL in general (for Sea Islands and previous, including R600)?
    I know the story about canceling fglrx-kernelside and bring it to floss. So I take your hint / link as: HSA is not implemented in fglrx, oss has a lead as planned by bringing kernel-side of fglrx into the oss driver
    However I would like to have a reliable source / confirmation from a dev or a link that actually states this information explicetly.

    Leave a comment:


  • asdfblah
    replied
    Originally posted by tomtomme View Post
    is this even implemented in fglrx? I think windows catalyst has hsa, but on linux with fglrx? does somebody know if oss has a lead here?
    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite


    How does this benefit other architectures, and OpenCL in general (for Sea Islands and previous, including R600)?

    Leave a comment:


  • pixo
    replied
    Just a few questions:
    Does that mean that until now for HSA hardware you had to move data between main mem and gpu dedicated mem?
    If so will these patches be used by the radeon driver to stop this to save bandwith?
    Would it be possible to write 3D engine in HSAIL and will there be benefit other than being portable to all HSA hw (kaveri and next gen phones, maybe PS4 and xbox1)?

    Leave a comment:


  • tomtomme
    replied
    Originally posted by nigeil View Post
    Very excited to hear about this release. Glad to see AMD making strides, however slow they are in coming, to push this support out to the open source community. Hopefully we can put it to good use; I was waiting to take advantage of this in some of my own scientific modelling software.
    is this even implemented in fglrx? I think windows catalyst has hsa, but on linux with fglrx? does somebody know if oss has a lead here?

    Leave a comment:


  • stiiixy
    replied
    Originally posted by Kivada View Post
    If you're in an uncivilized country like the US as I am save your cash for food and bills. At this point it's probably better to wait for the next gen version with mobos sporting DDR4 as DDR4 hardware was all the rage at Computex a few weeks ago. With those maybe we'll also get on mobo GDDR5 for some real fun with memory bandwidth...
    GDD5 mounted where on the hardware? On-die, the motherboard or maybe socketed like in the old FPU days? You sound like you saw or heard something about it. Now spill the beans!

    And when's the official DDR4 launch? I heard it was 'real soon now', but it's been like that for a couple years now. Coincidentally, about as long as the DDR3 price inflation occured.

    EDIT: Oh, and I think 'dynamic' instead of 'uncivilised' will will get you a few less scowls from the Muhrican's =D

    Leave a comment:


  • Kivada
    replied
    Originally posted by Adarion View Post
    Oh wow. So HSA actually becomes reality. It seems somewhat unreal to me, but a lot of things do since I dropped out of work into unemployment last week (after 7 months of 70... 100 h/week workload everything seems strange when you suddenly stop).
    I guess this might be an underrated start for something much larger. Well, future will show. Thanks to the developer team for all the effort and AMD in general for providing specs and free drivers.
    Time for me to show my support by buying another APU (while money still lasts ).
    If you're in an uncivilized country like the US as I am save your cash for food and bills. At this point it's probably better to wait for the next gen version with mobos sporting DDR4 as DDR4 hardware was all the rage at Computex a few weeks ago. With those maybe we'll also get on mobo GDDR5 for some real fun with memory bandwidth...

    Leave a comment:


  • Adarion
    replied
    Oh wow. So HSA actually becomes reality. It seems somewhat unreal to me, but a lot of things do since I dropped out of work into unemployment last week (after 7 months of 70... 100 h/week workload everything seems strange when you suddenly stop).
    I guess this might be an underrated start for something much larger. Well, future will show. Thanks to the developer team for all the effort and AMD in general for providing specs and free drivers.
    Time for me to show my support by buying another APU (while money still lasts ).

    Leave a comment:


  • stiiixy
    replied
    I was beginning to seriously consider Intel gear for a single workstation. I haven't had one since my P3-1200, which I keep as my sexy little over-sized firewall, wewts...oh wait I have an Intel ivy7 laptop, derp (with AMD GFX =)

    But this newspiece and subsequent user posts brought me back. Thank you to all and sundry!

    Shit's crackin'!

    Leave a comment:


  • rvdboom
    replied
    Questions to Mr Bridgman : what are the next step after the kernel drive for comprehensive HSA support? Will there be patches to compilers like GCC or LLVM? What modifications would be needed for existing codes to make use of HSA?
    Probably to wide a question but let's try...

    Leave a comment:


  • Kivada
    replied
    Originally posted by yoshi314 View Post
    In layman's terms, aside from compute and clustering - it sounds as it might speed up data transfers between cpu and gpu ? Is that the benefit from typical user, or is there none?
    There would be tons, every task that currently bogs down multiple CPU cores would greatly bennifit from this, the other posted techs that they are working on would vastly improve Java performance among other tasks. HSA is not just for AMD hardware, pretty much all of the major ARM players are involved as well, it's made to allow the CPU, GPU and DSP all work together, which will be comming to a cellphone near you soon enough.

    The single core IPC of the CPU has pretty much hit a wall, even on the Intel side, there aren't large gains every other version like the old days and most tasks that can be broken up to multiple threads don't gain a whole lot from a handful of general purpose CPU cores, but make huge gains when broken up into hundreds of pieces and sent off to the GPU.

    What kills performance of current GPGPU implementations is the lack of unified memory and cache coherency between the CPU and GPU. What this means is that the data would have to be copied into main system memory, the CPU would have to decide to send it out to the GPU, it's have to be sent across the PCIe bus to the GPU memory, then worked on and placed back into the GPU memory, then sent back across the PCIe to the CPU memory, then the CPU has to check it, all of this adds latency to the process blowing away allot of the gains of processing the data on the GPU limiting it to only non real time tasks like Bitcoin mining.

    One big gain for end users would be in GPU based physics in games, not just crappy non-persistent eye candy only ones like you see with PhysX, a game engine written to make use of the GPU in an APU would allow for all kinds of fun things to be done to the game environment, think a war game where you have a fully destructible city and more realistic(or not) hit effects on character's bodies.

    Got any multimedia task? The GPU would absolutely destroy any CPU on the market with ease in these tasks. Think how VDPAU/VA-API help with video playback, apply that to editing and transcoding file, which is a very time consuming task, especially as we move to 4K and eventually 8K video.

    Leave a comment:

Working...
X