Language Selection

English French German Italian Portuguese Spanish

Graphics/Benchmarks

Kernel and Graphics: LTS, NVIDIA, Mir 0.28, RADV

Filed under
Graphics/Benchmarks
Linux
  • Linux LTS Kernel Support Extended to 6 Years

    Despite being a free and open source OS, Linux has one of the quickest development cycles. Every 70 days, a major new version of the Linux kernel rolls out. This deprecates any older versions except for the few Long Term Support releases. In a sign of changing times, the Linux kernel lifetime is getting a major boost. From now on, the Linux LTS kernel will be supported for a whole 6 years. Jumping up to 6 years is a tripling of the current 2 year support period. This is especially goods news for Android.

  • A Reverse-Engineered Tegra Video Decode Driver Steps Closer To Mainline
  • NVIDIA Releases Linux Graphics Debugger 2.2

    NVIDIA has today released an updated version of their Linux Graphics Debugger to help game/application developers in analyzing issues and performance problems around OpenGL 4.x on GeForce/Quadro GPUs.

  • RADV Vulkan Driver Gets An On-Disk Shader Cache

    The RADV Radeon Vulkan driver in Mesa now supports an on-disk shader cache.

    Timothy Arceri working for Valve who previously spearheaded the RadeonSI OpenGL on-disk shader cache support has added a similar on-disk shader cache for the RADV Vulkan driver. Vulkan supports the concept of a pipeline cache for reusing cached objects between pipelines and runs of a game/application.

  • Another Minor Performance Optimization For RADV

    While Timothy Arceri working for Valve was busy wiring up an on-disk shader cache for RADV, Samuel Pitoiset working for this gaming giant has been tackling some additional optimizations.

  • Mir 0.28 Arrives As A Late Addition To Ubuntu 17.10

    Not that Mir is playing a pivotal role in Ubuntu 17.10 now that the transition has occurred to GNOME Shell and Wayland for this release, the new Mir release should make it into the archive. Mir 1.0 was expected up until yesterday when it was announced Mir 1.0 would be released as Mir 0.28.

  • The RADV Vulkan driver for AMD GPUs now has a shader cache in Mesa, plus more Mesa news

Linux Foundation, Kernel, and Graphics

Filed under
Graphics/Benchmarks
Linux
  • A ZSTD-Compressed Linux Kernel Could Be Up Next

    Nick Terrell of Facebook is proposing support for ZSTD-compressed kernel and ramdisk images. This would add to the list of other algorithms already available like Gzip, XZ, LZ4, and others for dealing with kernel images at boot time.

  • Chasing Grace: A New Documentary Series about Women in Tech

    After hearing several women in tech, smart women with bright futures, talk about leaving their jobs, Jennifer Cloer, Founder/Lead Consultant, reTHINKit PR, decided to launch the “Chasing Grace Project,” a six-episode documentary series about women in tech. The trailer debuted at the recent Linux Foundation Diversity Empowerment Summit in LA.

    “A young, very talented female programmer recently told me: ‘I don’t want to leave tech but after a year into my first job, I’m considering it,’” said Cloer. So she asked herself, “What can I do to help”

  • OVR_multiview Extension Completed For More Efficient OpenGL VR

    The OVR_multiview OpenGL Extension developed via the OpenVR initiative has been around for several months in an incomplete form for allowing more efficient virtual reality (VR) rendering while now the extension is complete.

  • More Than 100 More AMDGPU DC Patches Line Up Ahead Of Linux 4.15

    AMDGPU DC is expected for Linux 4.15 assuming Linus Torvalds has no objections to merging the code. We hope it won't, but the code-base for this new AMD display code is outright massive at more than 120,000 lines of code over hundreds of patches. Today another 103 new patches were published.

  • Initial Gallium3D VC5 Driver Merged Into Mesa

    The initial "VC5" Gallium3D driver for next-generation Broadcom graphics hardware has been merged into mainline Mesa.

  • Linux 4.15 Will Finally Graduate Intel "Coffee Lake" Graphics Out Of Alpha Support

    Another set of Intel Direct Rendering Manager (DRM) driver updates were mailed in to DRM-Next today for the eventual Linux 4.15 kernel cycle.

NVIDIA 387.12 Vulkan vs. OpenGL Performance Across Multiple CPUs

Filed under
Graphics/Benchmarks

Published earlier this week was the Intel Core i3, i5, i7 With NVIDIA vs. AMD Radeon For Linux Gaming results that are quite interesting while in this article is looking at the OpenGL vs. Vulkan Linux gaming performance using NVIDIA's first-rate binary driver while also doing this graphics API/renderer comparison across the Intel Coffeelake processors from low-end to high-end.

Read more

Graphics: Mesa, Mir, and Phoronix Test Suite

Filed under
Graphics/Benchmarks
  • Mesa Is Up To About 8,000 Commits This Year, 2.2 Million Lines

    Now being into Q4, I've been meaning to run some fresh Mesa Git development statistics to see how this year is pacing for this important piece of the open-source graphics ecosystem and Linux desktop.

    As of this morning, Mesa's Git tree is made up of 5,633 files representing a total of 2,243,544 lines. This came about over 96,443 commits from more than 800 different authors. Mesa is still seeing on average about 13 commits per day.

  • Mir 1.0 Is Ready For Release

    There is just one week to go until the Ubuntu 17.10 "Artful Aardvark" release but it looks like the Mir 1.0 release will still happen in time.

    It's been the goal of the remaining Mir developers to release version 1.0 during the Artful cycle. As of a few weeks back they were still aiming for that goal and to pursue a feature freeze exception. As of yesterday, all blocker bugs have been cleared for releasing Mir 1.0.

  • MESA_tile_raster_order Added To The OpenGL Registry

    The new OpenGL extension MESA_tile_raster_order proposed by Eric Anholt at Broadcom has now been merged to the Khronos registry.

  • Phoronix Test Suite 7.6-Alvdal Milestone 1 Released

Graphics: Vega 10 On RadeonSI, Vulkan 1.0.62

Filed under
Graphics/Benchmarks
  • Testing Primitive Binning With Vega 10 On RadeonSI

    One month back Marek Olšák landed support in the RadeonSI Gallium3D driver for primitive binning with Vega 10 GPUs but now that feature is likely to be disabled by default.

    Marek is moving ahead now to disable primitive binning by default for Vega 10 GPUs but to keep it enabled for upcoming Raven Ridge APUs. He explained in the proposed patch, "Our driver implementation is known to decrease performance for some tests, but we don't know if any apps and benchmarks (e.g. those tested by Phoronix) are affected. This disables the feature just to be safe."

  • Vulkan 1.0.62 Adds A New AMD Extension

    Vulkan 1.0.62 is now available as the latest updated specification for this high performance graphics and compute API.

    Vulkan 1.0.62 is mostly comprised of the usual documentation fixes and other clarifications. Nothing really too notable on that front with Vulkan 1.0.62.

Latest Open Access LWN: Fedora, Linux Kernel, and Graphics

Filed under
Graphics/Benchmarks
Linux
Red Hat
  • Fedora's foundations meet proprietary drivers

    The Fedora project's four "foundations" are named "Freedom", "Friends", "Features", and "First". Among other things, they commit the project to being firmly within the free-software camp ("we believe that advancing software and content freedom is a central goal for the Fedora Project, and that we should accomplish that goal through the use of the software and content we promote") and to providing leading-edge software, including current kernels. Given that the kernel project, too, is focused on free software, it is interesting to see a call within the Fedora community to hold back on kernel updates in order to be able to support a proprietary driver.

    On September 5, Fedora kernel maintainer Laura Abbott announced that the just-released 4.13 kernel would be built for the (in-development) Fedora 27 release, and that it would eventually find its way into the Fedora 25 and 26 releases as well. That is all in line with how Fedora generally operates; new kernels are pushed out to all supported releases in relatively short order. Running current kernels by default is clearly a feature that many Fedora users find useful.

    More recently, though, James Hogarth noted that the NVIDIA proprietary driver did not work with the 4.13 kernel. This kind of breakage is not all that unusual. While the user-space ABI must be preserved, the kernel project defends its right to change internal interfaces at any time. Any problems that out-of-tree code experiences as a result of such changes is deemed to be part of the cost of staying out of the mainline. There is little sympathy for those who have to deal with such issues, and none at all if the out-of-tree code in question is proprietary. Community-oriented projects like Fedora usually take a similar attitude, refusing to slow down for the sake of proprietary code.

  • Notes from the LPC tracing microconference

    The "tracing and BPF" microconference was held on the final day of the 2017 Linux Plumbers Conference; it covered a number of topics relevant to heavy users of kernel and user-space tracing. Read on for a summary of a number of those discussions on topics like BPF introspection, stack traces, kprobes, uprobes, and the Common Trace Format.

    Unfortunately, your editor had to leave the session before it reached its end, so this article does not reflect all of the topics discussed there. For those who are interested, this Etherpad instance contains notes taken by participants at the session.

  • An update on live kernel patching

    In the refereed track at the 2017 Linux Plumbers Conference (LPC), Jiri Kosina gave an update on the status and plans for the live kernel patching feature. It is a feature that has a long history—pre-dating Linux itself—and has had a multi-year path into the kernel. Kosina reviewed that history, while also looking at some of the limitations and missing features for live patching.

    The first question that gets asked about patching a running kernel is "why?", he said. That question gets asked in the comments on LWN articles and elsewhere. The main driver of the feature is the high cost of downtime in data centers. That leads data center operators to plan outages many months in advance to reduce the cost; but in the case of a zero-day vulnerability, that time is not available. Live kernel patching is targeted at making small security fixes as a stopgap measure until the kernel can be updated during a less-hurried, planned outage. It is not meant for replacing the kernel bit by bit over time, but as an emergency measure when the kernel is vulnerable.

  • Safety-critical realtime with Linux

    Doing realtime processing with a general-purpose operating-system like Linux can be a challenge by itself, but safety-critical realtime processing ups the ante considerably. During a session at Open Source Summit North America, Wolfgang Mauerer discussed the difficulties involved in this kind of work and what Linux has to offer.

    Realtime processing, as many have said, is not synonymous with "real fast". It is, instead, focused on deterministic response time and repeatable results. Getting there involves quantifying the worst-case scenario and being prepared to handle it — a 99% success rate is not good enough. The emphasis on worst-case performance is at the core of the difference with performance-oriented processing, which uses caches, lookahead algorithms, pipelines, and more to optimize the average case.

  • A memory allocation API for graphics devices

    At last year's X.Org Developers Conference (XDC), James Jones began the process of coming up with an API for allocating memory so that it is accessible to multiple different graphics devices in a system (e.g. GPUs, hardware compositors, video decoders, display hardware, cameras, etc.). At XDC 2017 in Mountain View, CA, he was back to update attendees on the progress that has been made. He has a prototype in progress, but there is plenty more to do, including working out some of the problems he has encountered along the way.

    Jones has been at NVIDIA for 13 years and has been working on this problem in various forms for most of that time, he said. Allocating buffers and passing them around between multiple drivers is a complicated problem. The allocator will sit in the same place as the Generic Buffer Management (GBM) component is today; it will be used both by applications and by various user-space driver components. The allocator will support both vendor-agnostic (e.g. Android ION) and vendor-specific back-ends, as well as combinations of the two.

Linux Kernel and Graphics: Jprobes, Performance Analysis in Linux, AMDGPU, UHD Graphics and Vulkan

Filed under
Graphics/Benchmarks
Linux
  • An end to jprobes

    "Jprobes" are an ancient kernel mechanism used to trace entry into kernel functions; they were described in this 2005 LWN article.

  • Performance analysis in Linux (continued)

    This blog post is based on the talk I gave at the Open Source Summit North America 2017 in Los Angeles. Let me start by thanking my employer Collabora, for sponsoring my trip to LA.

    Last time I wrote about Performance Assessment, I discussed how an apparently naive code snippet can hide major performance drawbacks. In that example, the issue was caused by the randomness of the conditional branch direction, triggered by our unsorted vector, which really confused the Branch Predictor inside the processor.

  • AMDGPU's DC Gets More Cleanups Ahead Of Linux 4.15

    It's indeed looking like the AMDGPU DC display code stack will finally be pulled for the Linux 4.15 merge window, assuming Linus Torvalds has no issues with it in a few weeks.

    Alex Deucher of AMD today sent in a secondary AMDGPU DC update for staging alongside DRM-Next of this new display code.

  • Running OpenCL With Intel UHD Graphics On Coffeelake Under Ubuntu Linux

    After running some basic OpenCL/Vulkan UHD Graphics tests yesterday using the brand new Core i7 8700K "Coffee Lake" processor, I next ventured into OpenCL computing with the UHD Graphics using Intel's open-source Beignet CL implementation.

  • VK_KHR_maintenance2 Lands For RADV, Other Improvements

    VK_KHR_maintenance2 is the latest extension supported by this open-source Radeon Vulkan driver. VK_KHR_maintenance2 was added in Vulkan 1.0.61 last month as various changes that were previously left out of Vulkan.

  • Intel UHD Graphics 630 "Coffee Lake" On Linux

    This morning I delivered the initial Linux processor benchmarks of the Core i7 8700K and Core i5 8400 for the just-launched "Coffee Lake" desktop processors. With these Intel "Gen 8" processors, the integrated "HD Graphics" from Kabylake have been rebranded to "UHD Graphics". While there wasn't any real changes architecturally to the graphics hardware, right now the Linux support isn't quite out-of-the-box.

  •  

Kernel, the Linux Foundation, and Latest Benchmarks

Filed under
Graphics/Benchmarks
Linux
  • Stable kernels 4.13.5, 4.9.53, 4.4.90, and 3.18.73
  • The Linux Foundation Announces 15 New Silver Members

    The Linux Foundation, the nonprofit organization enabling mass innovation through open source, announced that 15 new organizations have joined the Foundation as Silver members. Linux Foundation members help support development of the greatest shared technology resources in history, while accelerating their own innovation through open source leadership and participation.

  • More AMDGPU DC + RadeonSI/RADV vs. NVIDIA Linux OpenGL+Vulkan Benchmarks

    Last week I published the Radeon RX Vega Performance With Mesa 17.3-dev + LLVM 6 + drm-next-4.15-dc article offering a fresh look at the RX Vega 56/64 Linux performance using the new AMDGPU DC code that's likely to be merged in Linux 4.15. As well, the latest Mesa 17.3-dev Git code built against LLVM 6.0 SVN while all compared to the latest NVIDIA driver. Out of that article came some premium requests to see a larger comparison, so here that is.

  • Intel Core i5 8400 Linux Performance

    Today marks the embargo expiry for reviews on Intel's new Coffeelake desktop processors. While a CPU refresh may not normally be too exciting, thanks to the pressure from AMD with their Ryzen processors pushing core counts higher, Intel is now upping the core counts in their desktop CPUs. Today we will be featuring Linux benchmarks of the Core i5 8400 and Core i7 8700K while this article is focusing on the i5-8400: a six-core Core i5!

Intel Core i7 8700K Linux Benchmarks

Filed under
Graphics/Benchmarks

2017 has been an interesting year for processors with AMD's long awaited introduction of the Zen-based Ryzen / Threadripper / EPYC processors, Intel's Core X-Series processors for high-end desktops, the Xeon Scalable processor family introduction, and now the launch of Coffee Lake as a "Kaby Lake Refresh" step before the Cannonlake desktop processors expected in 2018. While another 14nm CPU, Coffee Lake is interesting is that Intel has now upped their desktop core counts in response to Ryzen. With the Core i7 series is now six cores plus Hyper Threading, compared to 4 cores plus HT with previous i7 models. The Core i5 CPUs are also now six core but sans Hyper Threading and there is also the just-published Core i5 8400 Linux benchmarks. This article serves as our first look at the Coffee Lake Core i7 CPUs in the form of the 8700K.

Read more

Graphics: X.Org Server 1.19.4 and RADV

Filed under
Graphics/Benchmarks
Syndicate content

More in Tux Machines

10 Reasons Why I Switched To Telegram Messenger

Whatsapp may be the best player in the game when it comes to instant messaging apps, but Telegram Messenger is the entire game itself. Read
more

How To Install Wine And Run Windows Apps In Linux

​All kinds of software are currently available on Linux but every now and then, there is that Windows software or Game which is not available or has no equivalent on Linux Wine makes it possible to run those Windows programs and Games on your Linux desktop. So let’s look at how to install Wine on Linux and run Windows apps on Linux desktop. Read
more

OSS Leftovers

Today in Techrights