Last week marked the release of Intel's Beignet 1.3, their open-source project implementing OpenCL acceleration atop modern CPUs with HD/Iris Graphics. Significant with Beignet 1.3 is that they've finally implemented OpenCL 2.0 support! OpenCL 2.0 is now available for Skylake hardware and newer. Beignet 1.3 also has other new features, runtime improvements, LLVM 3.9 support, new extensions, and much more. Thus time for some benchmarking of this new Beignet release.
LLVM Clang 4.0 is set to be released in February while GCC 7 will be released as stable in March~April. For those curious how both compilers are currently performing, here is our latest installment of GCC vs. LLVM Clang benchmarking on Linux x86_64.
With GCC7 feature development ending, this week I conducted some benchmarks of the latest GCC 7 snapshot against that of the past three major release series of the GNU Compiler Collection: 6.3.0, 5.4.0, and 4.9.4. All tests were done on Ubuntu Linux x86_64 with an Intel Core i7 6800K processor.
The twelfth year is now in the books at Phoronix. In 2016 on Phoronix there were 3,336 original news articles and 248 featured multi-page articles and Linux hardware reviews. That puts our total now at more than 21.3k news articles and 3.3k Linux hardware reviews and other featured articles. Happy New Year to all and 2017 will hopefully be even better.
Given this weekend's release of Darktable 2.2 as a big upgrade to this open-source RAW photo workflow software, here are some fresh benchmarks of NVIDIA GeForce and AMD Radeon graphics cards under Linux when making use of the program's OpenCL support, which did see some improvements during this v2.2 cycle.
As some more end-of-the-year Linux benchmarks, here are OpenCL Darktable and Blender benchmarks when testing on thirteen different NVIDIA GeForce graphics cards.
Earlier this week I published some GCC 5.4 vs. GCC 6.2 vs. GCC 7.0 SVN development benchmarks with a Core i7 6800K Broadwell-E system. For those curious how the LLVM Clang compiler stack is comparing, here are some tests on the same system when running fresh benchmarks of LLVM Clang 3.9 as well as LLVM Clang 4.0 SVN.
With now having netperf in the Phoronix Test Suite as well as iperf3 for the latest open-source benchmarks in our automated cross-platform benchmarking framework, I couldn't help but to run some networking benchmarks on a system when trying out a few different Linux distributions and BSDs to see how the performance compares. The operating systems ran with these networking benchmarks included Debian 8.6, Ubuntu 16.10, Clear Linux 12020, CentOS 7, and Fedora 25. The BSDs tested for this comparison were FreeBSD 11.0 and DragonFlyBSD 4.6.1.
With the GCC 7 compiler having entered its stage three, feature development is basically over so it's a great time to begin running more benchmarks of this big compiler update that will be officially released as GCC 7.1.0 in early 2017. Up today are benchmarks of the latest GCC 7.0 development snapshot compared to GCC 6.2 and GCC 5.4 on an Intel Core i7 6800K Broadwell-E system running Ubuntu 16.10.
With the holiday season in full swing, whether you are just a casual photographer or professional, Darktable is easily one of the best photography workflow applications and it's free software! Darktable has offered OpenCL acceleration for providing faster performance on GPUs and with the imminent Darktable 2.2 release there is even better OpenCL results. For those curious about the OpenCL performance of Darktable, I've done some Darktable 2.2-RC1 benchmarks on a variety of NVIDIA GeForce and AMD Radeon graphics cards under Ubuntu Linux.
For the latest benchmarking off MSI's Cubi 2 with Core i5 Kaby Lake CPU are some GCC and LLVM Clang compiler benchmarks on Intel's Clear Linux distribution.
Following last week's benchmarks of the GeForce GTX 1050 and GTX 1050 Ti has been a request to see some fresh Blender benchmarks with CUDA acceleration of the Pascal line-up. Now having more time with these latest GTX 1000 series cards, here are such benchmarks.
Earlier this week I published Linux 3.9 through Linux 4.9 kernel benchmarks looking at the raw performance of various subsystems when testing each of the major kernel releases as far back as this Core i7 Haswell system was supported. From that same system, today is a look at testing the kernels going back to Linux 3.11 when Haswell graphics support was first in good shape for this Core i7 4790K box while looking at the raw power consumption and performance-per-Watt for these 19 major kernel releases.
With the in-development Linux 4.9 kernel showing signs of some performance improvements, I've gone ahead and tested the last 21 major kernel releases on the same system. From Linux 3.9 to Linux 4.9, each of the major kernel releases was tested from the same Intel Core i7 desktop with a variety of benchmarks.
This weekend was the release of Linux 4.9-rc1 to mark the end of the 4.9 kernel merge window. As such, here's our usual feature overview recapping all of the changes to Linux 4.9 that have us excited about the next version of this open-source kernel.
With this month's Ubuntu 16.10 "Yakkety Yak" release, Unity 7 on the X.Org Server is used by default on the desktop while an alternate Unity 8 + Mir session is installed by default, but while it's available is not used unless logging into this newer desktop stack. Here's my few minutes of trying it out this morning with all of the latest Ubuntu 16.10 packages.
If all goes well, GNOME 3.22 will be officially released tomorrow, 21 September. Here is a recap of some of the new features and improvements made over this past six month development cycle plus some screenshots of the near-final desktop that will power the upcoming Fedora 25 Workstation.
Following yesterday's GCC 5 vs. 6 vs. early 7 benchmarks, to no surprise LLVM's Clang compiler was brought up in the comments. I had already been running some fresh LLVM Clang benchmarks on this same Intel Xeon system and have those results to share now with Clang 3.8 and the newly-released Clang 3.9.
While GCC 7 is still under heavy development and the GCC 7.1 stable release will not come until a few months into 2017, here are some early benchmarks of GCC 7.0 compared to GCC 6.2 and GCC 5.4 on an Ubuntu Linux x86_64 system.
For those interested in C/C++ compiler performance, for some fun numbers to dive into this weekend are LLVM Clang vs. GCC benchmarks atop FreeBSD 11.0 RC1 AMD64 on an Intel Xeon Haswell system.
Given the underlying work that's been happening in the CPUFreq/scheduler area and the introduce of the new Schedutil CPUFreq governor, I decided to run some fresh performance benchmarks of P-State and CPUFreq with the different governor options when testing from a Linux 4.8 Git kernel atop the current Fedora 25 development packages and using a Core i5 Skylake processor.
Today marks the closure of the Linux 4.8 kernel merge window so as usual here is our recap of all the features we've been monitoring over the past two weeks. Among the highlights for Linux 4.8 are AMD GPU OverDrive overclocking, initial NVIDIA Pascal support, a new ARM Mali display driver, mainline support for the Raspberry Pi 3 BCM SoC, HDMI CEC support, big reworks to Btrfs and XFS file-system code, and a number of new security features, among other changes.
Here is a look at the performance of the Blender 3D modeling/creation software with its Cycles Engine when making use of NVIDIA's CUDA API for GPU acceleration. Tests for this initial comparison include NVIDIA GeForce GTX 1000 "Pascal" and GTX 900 "Maxwell" graphics cards.
What does it mean when developers behind one of the world's most popular desktop environments decide to jump into the deep end and fork a distribution? Depending on who you ask you'll hear madness, excellence, confusion, and excitement as onlookers figure out the exact nature of a new breed of beast and guess what it will do.
After a very exciting past two weeks, the merge window for Linux 4.7 is expected to close today. This was an action-packed merge window with a ton of new code being introduced. While I've already written dozens of posts on Phoronix about the changes that got me excited, here's my usual kernel feature overview. Here's a look at what's coming for Linux 4.7.
With the in-development Linux 4.7 kernel there is a new CPUFreq governor that leverages the kernel's scheduler utilization data in an attempt to make better decisions about adjusting the CPU's frequency / performance state. Here are some benchmarks of that new CPUFreq governor, Schedutil, compared to the other CPUFreq governors as well as the Intel P-State CPU frequency scaling driver.
After carrying out the recent GCC 4.9 vs. 5.3 vs. 6.1 compiler benchmarks for looking at the GNU Compiler Collection performance over the past three years on the same Linux x86_64 system, I then loaded up a development snapshot of the LLVM 3.9 SVN compiler to see how these two dominant compilers are competing on the performance front for C/C++ programs.
Given the recent stable release of GCC 6 (GCC v6.1.0), here are some fresh compiler benchmarks on an Intel Debian x86_64 system when comparing the GCC 4.9.3, GCC 5.3.0, and GCC 6.1.0 compiler releases.
One of the most recurring requests this week from Phoronix readers were for doing some ZFS file-system tests on Ubuntu 16.04 LTS. Here are some basic results using a single SSD.
Some Phoronix readers have been requesting fresh tests of OpenGL graphics/gaming performance on Ubuntu 16.04 LTS with its different desktop environment options. For some brief results to share this Sunday, here are some Intel Skylake numbers when running Ubuntu 16.04 and testing out Unity, Xfce, KDE Plasma, LXDE, GNOME, MATE, and Openbox.
537 software articles published on Phoronix.