gpu News - Page 15 of 30 - CNX Software

March 27, 2019 by Jean-Luc Aufranc (CNXSoft) - 20 Comments

ODROID-N2 GPU Drivers, Linux 5.0, and Impressive glmarks-es2 Score

ODROID-N2 was announced last February for $63 (2GB RAM), and $79 (4GB RAM), but Hardkernel was not quite ready to take orders at the time. One of the good news is that the 4GB RAM is now available for pre-order with shipping scheduled to start on April 3. Another good news is on the software side with Hardkernel having released the userland Mali-G52 Wayland driver. It does not work well with Linux 4.9 due to incomplete DRM implementation, but it goes work with Linux 5.0 plus some modifications as further discussed in the aforelinked forum thread. The screenshot above, courtesy of odroid forum member memeka , shows ODROID-N2 running Ubuntu 18.04 + Gnome3 + Linux 5.0 on top of Wayland with GPU drivers providing acceleration as shown by glmark2-es2-wayland test program. The benchmark results are pretty impressive:

=======================================================
    glmark2 2014.03+git20150611.fa71af2d
=======================================================
    OpenGL Information
    GL_VENDOR:     ARM
    GL_RENDERER:   Mali-G52
    GL_VERSION:    OpenGL ES 3.2 v1.r16p0-01rel0.2943fc4ef9657d91ee32c9a58dec6cd2
=======================================================
[build] use-vbo=false: FPS: 961 FrameTime: 1.041 ms
[build] use-vbo=true: FPS: 1592 FrameTime: 0.628 ms
[texture] texture-filter=nearest: FPS: 1491 FrameTime: 0.671 ms
[texture] texture-filter=linear: FPS: 1477 FrameTime: 0.677 ms
[texture] texture-filter=mipmap: FPS: 1524 FrameTime: 0.656 ms
[shading] shading=gouraud: FPS: 1151 FrameTime: 0.869 ms
[shading] shading=blinn-phong-inf: FPS: 1215 FrameTime: 0.823 ms
[shading] shading=phong: FPS: 1043 FrameTime: 0.959 ms
[shading] shading=cel: FPS: 1126 FrameTime: 0.888 ms
[bump] bump-render=high-poly: FPS: 514 FrameTime: 1.946 ms
[bump] bump-render=normals: FPS: 1976 FrameTime: 0.506 ms
[bump] bump-render=height: FPS: 1777 FrameTime: 0.563 ms
[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 1139 FrameTime: 0.878 ms
[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 383 FrameTime: 2.611 ms
[pulsar] light=false:quads=5:texture=false: FPS: 2096 FrameTime: 0.477 ms
[desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 389 FrameTime: 2.571 ms
[desktop] effect=shadow:windows=4: FPS: 788 FrameTime: 1.269 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 103 FrameTime: 9.709 ms
[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 129 FrameTime: 7.752 ms
[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 158 FrameTime: 6.329 ms
[ideas] speed=duration: FPS: 356 FrameTime: 2.809 ms
[jellyfish] <default>: FPS: 979 FrameTime: 1.021 ms
[terrain] <default>: FPS: 52 FrameTime: 19.231 ms
[shadow] <default>: FPS: 437 FrameTime: 2.288 ms
[refract] <default>: FPS: 88 FrameTime: 11.364 ms
[conditionals] fragment-steps=0:vertex-steps=0: FPS: 1769 FrameTime: 0.565 ms
[conditionals] fragment-steps=5:vertex-steps=0: FPS: 1769 FrameTime: 0.565 ms
[conditionals] fragment-steps=0:vertex-steps=5: FPS: 1853 FrameTime: 0.540 ms
[function] fragment-complexity=low:fragment-steps=5: FPS: 1783 FrameTime: 0.561 ms
[function] fragment-complexity=medium:fragment-steps=5: FPS: 1726 FrameTime: 0.579 ms
[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 1715 FrameTime: 0.583 ms
[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 1650 FrameTime: 0.606 ms
[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 1728 FrameTime: 0.579 ms
=======================================================
                                  glmark2 Score: 1119
=======================================================

=======================================================

glmark2 2014.03+git20150611.fa71af2d

=======================================================

OpenGL Information

GL_VENDOR: ARM

GL_RENDERER: Mali-G52

GL_VERSION: OpenGL ES 3.2 v1.r16p0-01rel0.2943fc4ef9657d91ee32c9a58dec6cd2

=======================================================

[build] use-vbo=false: FPS: 961 FrameTime: 1.041 ms

[build] use-vbo=true: FPS: 1592 FrameTime: 0.628 ms

[texture] texture-filter=nearest: FPS: 1491 FrameTime: 0.671 ms

[texture] texture-filter=linear: FPS: 1477 FrameTime: 0.677 ms

[texture] texture-filter=mipmap: FPS: 1524 FrameTime: 0.656 ms

[shading] shading=gouraud: FPS: 1151 FrameTime: 0.869 ms

[shading] shading=blinn-phong-inf: FPS: 1215 FrameTime: 0.823 ms

[shading] shading=phong: FPS: 1043 FrameTime: 0.959 ms

[shading] shading=cel: FPS: 1126 FrameTime: 0.888 ms

[bump] bump-render=high-poly: FPS: 514 FrameTime: 1.946 ms

[bump] bump-render=normals: FPS: 1976 FrameTime: 0.506 ms

[bump] bump-render=height: FPS: 1777 FrameTime: 0.563 ms

[effect2d] kernel=0,1,0;1,-4,1;0,1,0;: FPS: 1139 FrameTime: 0.878 ms

[effect2d] kernel=1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 383 FrameTime: 2.611 ms

[pulsar] light=false:quads=5:texture=false: FPS: 2096 FrameTime: 0.477 ms

[desktop] blur-radius=5:effect=blur:passes=1:separable=true:windows=4: FPS: 389 FrameTime: 2.571 ms

[desktop] effect=shadow:windows=4: FPS: 788 FrameTime: 1.269 ms

[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 103 FrameTime: 9.709 ms

[buffer] columns=200:interleave=false:update-dispersion=0.9:update-fraction=0.5:update-method=subdata: FPS: 129 FrameTime: 7.752 ms

[buffer] columns=200:interleave=true:update-dispersion=0.9:update-fraction=0.5:update-method=map: FPS: 158 FrameTime: 6.329 ms

[ideas] speed=duration: FPS: 356 FrameTime: 2.809 ms

[jellyfish] <default>: FPS: 979 FrameTime: 1.021 ms

[terrain] <default>: FPS: 52 FrameTime: 19.231 ms

[shadow] <default>: FPS: 437 FrameTime: 2.288 ms

[refract] <default>: FPS: 88 FrameTime: 11.364 ms

[conditionals] fragment-steps=0:vertex-steps=0: FPS: 1769 FrameTime: 0.565 ms

[conditionals] fragment-steps=5:vertex-steps=0: FPS: 1769 FrameTime: 0.565 ms

[conditionals] fragment-steps=0:vertex-steps=5: FPS: 1853 FrameTime: 0.540 ms

[function] fragment-complexity=low:fragment-steps=5: FPS: 1783 FrameTime: 0.561 ms

[function] fragment-complexity=medium:fragment-steps=5: FPS: 1726 FrameTime: 0.579 ms

[loop] fragment-loop=false:fragment-steps=5:vertex-steps=5: FPS: 1715 FrameTime: 0.583 ms

[loop] fragment-steps=5:fragment-uniform=false:vertex-steps=5: FPS: 1650 FrameTime: 0.606 ms

[loop] fragment-steps=5:fragment-uniform=true:vertex-steps=5: FPS: 1728 FrameTime: 0.579 ms

=======================================================

glmark2 Score: 1119

=======================================================

I’ve never seen such as high score (1,119 points) on Arm hardware. […]

March 15, 2019 by Jean-Luc Aufranc (CNXSoft) - 3 Comments

Turnip is an Open Source Vulkan Driver for Adreno GPU

Qualcomm Adreno GPUs have one of the best open source GPU driver for Arm SoC thanks to Freedreno driver. The driver relies on OpenGL ES API however, and nearly four of years ago, Khronos introduced the Vulkan API that aims to reduce CPU resources usage, and adds support multiple command buffers. The good news is there has been development of a Vulkan driver for Adreno 500 and 600 series GPU called Turnip, also referred to as Freedreno_vk (Freedreno Vulkan) in the code. The Turnip driver has been developed by Bas Nieuwenhuizen, a Site Reliability Engineer at Google who also happens to be RADV (Radeon Vulkan) lead developer, Chia-I Wu, a Google software engineer working on Android graphics, Chad Versace, Intel’s software engineer working on Linux OpenGL stack, and others as revealed by a recent merge into Mesa 19.1 slated to be released next quarter. Via Phoronix Jean-Luc Aufranc (CNXSoft)Jean-Luc started […]

December 26, 2018December 28, 2018 by Jean-Luc Aufranc (CNXSoft) - 7 Comments

Linux 4.20 Release – Main Changes, Arm and MIPS Architectures

After Greg K-H handling Linux 4.19 release, Linus Torvalds is back at the helm, and released Linux 4.20 just before Christmas: Let’s face it, last week wasn’t quite as quiet as I would have hoped for, but there really doesn’t seem to be any point to delay 4.20 because everybody is already taking a break. And it’s not like there are any known issues, it’s just that the shortlog below is a bit longer than I would have wished for. Nothing screams “oh, that’s scary”, though. And as part of the “everybody is already taking a break”, I can happily report that I already have quite a few early pull requests in my inbox. I encouraged people to get it over and done with, so that people can just relax over the year-end holidays. In fact, I probably won’t start pulling for a couple of days, but otherwise let’s just […]

August 18, 2018August 18, 2018 by Jean-Luc Aufranc (CNXSoft) - No Comments

Embedded Linux Conference Europe & OpenIoT Summit Europe 2018 Schedule

Embedded Linux Conference OpenIOT Summit Europe 2018

The Embedded Linux Conference & OpenIoT Summit 2018 took place in March of this year in the US, but the European version of the events are now planned to take place on October 21-24 in Edinburg, UK, and the schedule has already been released. So let’s make a virtual schedule to find out more about some of interesting subjects that are covered at the conferences. The conference and summit really only officially start on Monday 22, but there are a few talks on Sunday afternoon too. Sunday, October 21 13:30 – 15:15 – Tutorial: Introduction to Quantum Computing Using Qiskit – Ali Javadi-Abhari, IBM Qiskit is a comprehensive open-source tool for quantum computation. From simple demonstrations of quantum mechanical effects to complicated algorithms for solving problems in AI and chemistry, Qiskit allows users to build and run programs on quantum computers of today. Qiskit is built with modularity and extensibility […]

July 14, 2018September 28, 2021 by Jean-Luc Aufranc (CNXSoft) - 22 Comments

ROC-RK3328-CC Board Review, or the Case for Fast Storage and Adequate Power Supply

Firefly team from T-chip company has send me some of their Rockchip development boards, and we’ve already checked the provided boards and accessories, so today I’ll report my experience with one of the board: ROC-RK3328-CC also known as Renegade. I won’t test it with Android, since I have already reviewed RK3328 Android TV boxes such as Zidoo X7, and I’ve been told the team is hard at work with Android 8.1 SDK, so an Oreo image should be released in a few weeks/months. So I had initially planned to report my experience with one of the Linux images, then show how to install mainline Linux (currently 4.17) to the board, and reports what works. However, I encountered many issues, although likely not directly related to the board or its software support, so instead I’ll write about my experience getting started with the board, and list all the issues I had […]

June 1, 2018September 3, 2018 by Jean-Luc Aufranc (CNXSoft) - 9 Comments

Arm Announces Cortex-A76 CPU with Laptop-class Performance, Mali-G76 GPU, Mali-V76 8K VPU

Arm Cortex A75 based processors are only found in a few SoCs and devices, but Arm keeps on innovating, and they’ve now announced a new suite of of IP with Cortex-A76 CPU enabling 35 percent more performance, and Mali-G76 GPU with ML support and 30 percent higher efficiency and performance. SoC based on those new CPU and GPU IP will provide “laptop-class” performance, and the company also announced Arm Mali-V76 VPU with support for 8K video decoding and encoding. Arm Cortex A76 After Cortex A75, the Arm Cortex-A76 CPU is the second high performance processor core based on DynamIQ technology, and beside the 35 percent performance gain mentioned in the introduction, it also offers 40 percent improved efficiency, as well as delivers 4x compute performance improvements for AI/ML at the edge. Highlights of Cortex A76: Architecture – Armv8-A (Harvard) with Armv8.1, Armv8.2, Armv8.3 (LDAPR instructions only), cryptography and RAS extensions […]

March 14, 2018March 14, 2018 by Jean-Luc Aufranc (CNXSoft) - 4 Comments

Status of Embedded GPU Ecosystem – Linux/Mesa Upstream Support (ELC 2018 Video)

The Embedded Linux Confernce is on-going, and the Linux Foundation has been uploading videos about talks in a timely manner on YouTube. I checked out at RISC-V keynote yesterday, but today I’ve watched a talk by Robert Foss (his real name, not related to FOSS) from Collabora entitled “Progress in the Embedded GPU Ecosystem”, where he discusses open source software support in Linux/Mesa from companies and reverse-engineering support. The first part deals with the history of embedded GPU support, especially when it comes to company support. Intel was the first and offers very good support for their drivers, following by AMD who also is a good citizen. NVIDIA has the Nouveau driver but they did not really backed it up, and Tegra support is apparently sponsored by an aircraft supplier. Other companies have been slower to help, but Qualcomm has made progress since 2015 and now support all their hardware, […]

March 8, 2018 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

Vulkan 1.1 and SPIR-V 1.3 Specifications Released

The Khronos Group released Vulkan 1.0 specifications in 2015 as a successor of OpenGL ES, compatible with OpenGL ES 3.1 or greater capable GPU, and taking less CPU resources thank to – for instance – better use of multi-core processors with support for multiple command buffers that can be created in parallel. A year later, we saw Vulkan efficiency in a demo, since then most vendors have implemented a Vulkan driver for their compatible hardware across multiple operating systems, including Imagination Technologies which recently released Vulkan drivers for Linux. The Khronos Group has now released Vulkan 1.1 and the associated SPIR-V 1.3 language specifications. New functionalities in Vulkan 1.1: Protected Content – Restrict access or copying from resources used for rendering and display, secure playback and display of protected multimedia content Subgroup Operations – Efficient mechanisms that enable parallel shader invocations to communicate, wide variety of parallel computation models supported […]