Language Selection

English French German Italian Portuguese Spanish

NVIDIA graphics drivers to go multithreaded

Filed under
Software

techreport.com spoke recently with Ben de Waal, NVIDIA's Vice President of GPU software, and he revealed that NVIDIA has plans to produce multithreaded ForceWare graphics drivers for its GeForce graphics products.

Multithreading in the video driver should allow performance increases when running 3D games and applications on dual-core CPUs and multiprocessor PCs. De Waal estimated that dual-core processors could see performance boosts somewhere between five and 30% with these drivers.

Most imminent on the horizon right now is ForceWare release 75, which will bring a number of improvements for SLI performance and 64-bit Windows, among other things, but release 75 will not be multithreaded. The next major iteration of the driver, release 80, is slated to bring support for multiple threads. We may not see this version for a few months; NVIDIA hasn't given an exact timetable for the completion of release 80.

Out of curiosity, I asked de Waal why NVIDIA's drivers don't already take advantage of a second CPU. After all, the driver is a separate task from the application calling it, and Hyper-Threaded and SMP systems are rather common. He explained that drivers in Windows normally run synchronously with the applications making API calls, so that they must return an answer before the API call is complete. On top of that, Windows drivers run in kernel mode, so the OS isn't particularly amenable to multithreaded drivers. NVIDIA has apparently been working on multithreaded drivers for some time now, and they've found a way to fudge around the OS limitations.

De Waal cited several opportunities for driver performance gains with multithreading. Among them: vertex processing. He noted that NVIDIA's drivers currently do load balancing for vertex processing, offloading some work to the CPU when the GPU is busy. This sort of vertex processing load could be spun off into a separate thread and processed in parallel.

Some of the driver's other functions don't lend themselves so readily to parallel threading, so NVIDIA will use a combination of fully parallel threads and linear pipelining. We've seen the benefits of linear pipelining in our LAME audio encoding tests; this technique uses a simple buffering scheme to split work between two threads without creating the synchronization headaches of more parallel threading techniques.

Despite the apparent gains offered by multithreading, de Waal expressed some skepticism about the prospects for thread-level parallelism for CPUs. He was concerned that multithreaded games could blunt the impact of multithreaded graphics drivers, among other things.

Source.

More in Tux Machines

GNU/Linux on the Desktop Versus Proprietary Forms

  • Why I use a Mac computer, but an Android phone
    Yes, you could use a flavour of Linux on cheaper hardware, but then you trade the great Mac graphical interface with the ones available to Linux. You can fight me in the comments, but deep down you know I’m right. MacOS comes with Bash, and many of the tools those familiar with Linux would expect to have by default in their favourite distribution, including basics like “whois”, which aren’t installed in Windows by default.
  • Everything you knew about Chromebooks is wrong
    The original assumed vision of the Chromebook platform was a laptop and operating system capable of running only the Chrome web browser. You could do anything you wanted, as long as you wanted to stay on the web at all times. Today, the best new Chromebooks can runs apps from three additional operating systems. Not only do Chromebooks run apps, but they run more apps without dual- or multi-booting than any other computing platform. Chromebooks can run apps from Android, Linux and Windows concurrently in the same session.
  • Games, Tests and GitLab CI
    We are getting midterm of the GNOME 3.30 development cycle and many things already happened in the Games world. I will spare the user facing news for later as today I want to tell you about development features we desperatly needed as maintainers: tests and continuous integration. TL;DR: GLib, Meson, Flatpak and GitLab CI make writing and running tests super easy!

Graphics: Vulkan and Vega M

  • Vulkan Virgl Has Kicked Off For Supporting This Graphics/Compute API Within VMs
    Of the hundreds of projects for this year's Google Summer of Code, there are many interesting GSoC 2018 projects but one of those that I am most excited for is Vulkan-Virgl for getting this modern API supported with hardware acceleration by guest virtual machines. As implied by the name, this effort is based upon the Virgl project started by David Airlie and originally tasked with getting OpenGL acceleration to guest VMs using a fully open-source Linux driver stack. Virgl has been in good shape for a while now with OpenGL, while this summer the hope is to get the Vulkan API support going for opening up VMs to using this high-performance graphics and compute API.
  • AMDVLK Driver Lands Half-Float Additions, Many Other Improvements
    There's been another weekly-ish public code push to the AMDVLK open-source AMD Vulkan Linux driver stack and this time around it's heavy on feature work. There has been a fair amount of changes pertaining to half-float (FP16) support including support for the AMD_gpu_shader_half_float extension, prepping for VK_AMD_gpu_shader_half_float_fetch, FP16 interpolation intrinsics and register settings, and more.
  • Vega M Graphics On Intel Kabylake G CPUs Are Beginning To Work Under Linux
    We have been covering the Linux driver upbringing of "Vega M" for the Vega/Polaris graphics found in select newer Intel "Kabylake G" processors. The code is still in flight before it will work in all released versions of the Linux driver components, but for those willing to build the code or rely upon third party repositories, Vega M is now working on Linux. As I have covered in various past articles, the open-source driver support for Radeon Vega M is queued into DRM-Next for the upcoming Linux 4.18 kernel cycle, Mesa 18.1 albeit with new hardware I always recommend using the latest Git (current Mesa 18.2), and there are also binary GPU microcode files needed too.

Plasma 5.13 – Amazing Tux, How Sweet Plasma

Plasma 5.13 is (going to be) a very nice release. It builds on the solid foundation that is the LTS edition, and adds cool, smart touches. The emphasis is on seamless integration of elements, which is what separates professionals from amateurs. It’s all around how the WHOLE desktop behaves, and not individual programs in isolation. And Plasma is making great strides, offering a polished version of an already mature and handsome product, with extra focus on fonts, media and browser connectivity and good performance. There are some rough patches. Apart from the obvious beta issues, those goes without saying, KDE Connect ought to be a true multi-phone product, the network stack really needs to be spotless, and that means full Microsoft Windows inter-operability, Spectacle should allow for configurable shadows and alpha channel, and I want to see if the decorative backend has been cleaned up, i.e. can you search and install new themes and icons without encountering useless errors and inconsistencies. But all in all, I’m quite impressed. The changes are big and noticeable, and above all, meaningful. You don’t just get features for the sake of it, you get things that improve the quality and consistency of the desktop, that maximize fun and productivity, and there’s deep thought in orchestrating it all together. It ain’t just a random bunch of options that happen to work. I like seeing patterns in things, and I’m happy when there’s functional harmony. This spring season of distro testing hasn’t been fun, and Plasma 5.13 is balm for my weary wrists, so hurting from all that angry typing. More than worth a spin, and highly recommended. Full steam on, Tuxers. Read more Also: This week in Usability & Productivity, part 20

Sad News! Development Stopped for Korora and BackSlash Linux

It seems more and more small distributions are facing a had time. Recently we saw the crisis at Void Linux. Now we have two more small Linux distributions calling it quit, albeit temporarily. Read more