Software Linux Reviews & Articles

There have been 904 Linux hardware reviews and benchmark articles on Phoronix for software. Separately, check out our news section for related product news.

One of the interesting milestones this year in the compiler world was the ability with LLVM Clang 9.0 to compile Linux 5.3+ for x86_64 without needing any extra patches to either the kernel or the LLVM/Clang compiler. That initial support in Linux 5.3 was not without a few issues, but on Linux 5.5 the experience is in great shape with the stable Clang compiler.

Going back to the start of December with the Linux 5.5 merge window we have encountered several significant performance regressions. Over the weeks since we've reproduced the behavior on both Intel and AMD systems along with large and small CPUs. Following some holiday weekend bisecting fun, here is the cause at least partially for the Linux 5.5 slowdowns.

Last week were some benchmarks showing LLVM Clang hitting ~96% the performance of GCC using Intel Ice Lake while now for the recently released Zen2-based AMD Ryzen Threadripper 3960X we are seeing results where overall LLVM Clang is now at performance parity to GCC.

As part of our end-of-year benchmark comparisons, the latest results are looking at how the GNU Compiler Collection has evolved with the past five years of performance in testing GCC 5 through GCC 9 stable and the latest GCC 10 development compiler from the same system.

With 2019 quickly drawing to an end, I figured it would be interesting to see how the performance of Mozilla Firefox has been trending over the longer term. So for this article today is a look at the Firefox 57 through Firefox 71 stable performance plus tests of Firefox 72 beta and Firefox 73 alpha all from the same system and using a variety of browser benchmarks.

Linux 5.5-rc1 is on the way to mirrors and with that the Linux 5.5 merge window is now over. Here is a look at the lengthy set of changes and new features for this next Linux kernel that will debut as stable in early 2020.

Last week marked the release of Blender 2.81 with one of the shiny new features being the OptiX back-end for the Cycles engine to provide hardware-accelerated ray-tracing with NVIDIA RTX graphics processors. Long story short, OptiX is much faster for Blender than using NVIDIA's CUDA back-end -- which already was much faster than the OpenCL support within Blender. For your viewing pleasure today are benchmarks of 19 different graphics cards looking at the CUDA performance from Maxwell to Pascal to Turing and then for the RTX graphics cards also the OptiX performance.

Following the initial tests earlier this month from the disclosures of the JCC Erratum (Jump Conditional Code) that required updated Intel CPU microcode to address and on the same day the TSX Async Abort (TAA) vulnerability that required kernel mitigations to address, which I have run benchmarks of those CPU performance impacts individually, readers have requested tests looking at the current overall impact to the mitigations to date.

For those wondering if it's worthwhile for performance recompiling your key Linux binaries with the microarchitecture instruction set extensions and tuning for Ice Lake, here are some GCC compiler benchmarks looking at that impact for the Core i7 1065G7 on the Dell XPS 7390.

PHP 7.4 is due to be released next week as the annual major iteration to PHP7. Like we have seen through the PHP7 releases, while new features continue to be tacked on for this popular web-based programming language the performance has continued evolving. Here are the latest benchmarks of PHP 5.6 through PHP 7.4 while also looking at the PHP 8.0-dev performance that is in development on Git master.

Using Firefox 70 (including WebRender) and Google Chrome 78, here are our latest round of Linux web browser benchmarks tested on the Dell XPS Ice Lake laptop. Making this round of Linux browser benchmarking more interesting is also including power consumption and RAM usage metrics for the different browser benchmarks.

While this week we have posted a number of benchmarks on the JCC Erratum and its CPU microcode workaround that introduces new possible performance hits, also being announced this week as part of Intel's security disclosures was "Zombieload Variant Two" as the TSX Async Abort vulnerability that received same-day Linux kernel mitigations. I've been benchmarking the TAA mitigations to the Linux kernel since the moment they hit the public Git tree and here are those initial benchmark results on an Intel Cascade Lake server.

With yesterday's overview and benchmarks of Intel's Jump Conditional Code Erratum one of the areas where the performance impact of the updated CPU microcode exceeding Intel's 0~4% guidance was on the web browser performance. Now with more time having passed, here are more web browser benchmarks on both Chrome and Firefox while comparing the new CPU microcode release for the JCC Erratum compared to the previous release. Simply moving to this new CPU microcode does represent a significant hit to the web browser performance.

With the Linux 5.4 kernel set to be released in the next week or two, here is a look at the performance going back to the days of Linux 4.16 from early 2018. At least the Linux kernel continues picking up many new features as due to security mitigations and other factors the kernel performance continues trending lower.

With the new releases of Mozilla Firefox 70 and Google Chrome 78 here are fresh benchmarks of these web browsers with testing under Ubuntu Linux. Additionally, on the Firefox side looking at the performance with WebRender and compared to prior releases.

For those wondering if -- or how much -- of a performance impact mitigations still make regarding Spectre for Intel's long-awaited 10nm+ Ice Lake processors, here is the rundown on the mitigation state and the performance impact.

For those thinking of playing with Ubuntu 19.10's new experimental ZFS desktop install option in opting for using ZFS On Linux in place of EXT4 as the root file-system, here are some quick benchmarks looking at the out-of-the-box performance of ZFS/ZoL vs. EXT4 on Ubuntu 19.10 using a common NVMe solid-state drive.

One of the areas of performance I had been meaning to look more at following the recent AMD EPYC 7002 series launch was for database servers. With the original EPYC 7000 series performance, the performance came up short in competing with Intel Xeon CPUs, but for the EPYC Rome processors it ends up being a very different story. Given the launch last week of PostgreSQL 12, I've been trying out this new database server release on both EPYC and Xeon processors.

Complementing the 18-way NVIDIA GPU compute comparison from earlier this week with now having our hands on the RTX 2060/2070/2080 SUPER graphics cards, this round of NVIDIA Linux testing is looking at the Blender 2.80 and LuxCoreRender 2.1/2.2 performance for these popular rendering programs that offer CUDA acceleration.

With Intel seemingly ramping up work on their open-source OSPray portable ray-tracing engine now that they have pulled it under their oneAPI umbrella as part of a forthcoming rendering tool-kit, I figured it would be the latest interesting candidate for benchmarking of AMD EPYC 7742 vs. Intel Xeon Platinum 8280 performance. In addition, the Embree ray-tracing kernels are also being benchmarked as part of this performance comparison.

The Linux 5.4 merge window is set to end today with the release of Linux 5.4-rc1. With the major pull requests in, here is a look at the prominent changes and new features coming with Linux 5.4. As is standard practice, there will be about eight weekly release candidates of Linux 5.4 prior to officially releasing this kernel as stable in late November or potentially early December depending upon how the cycle plays out.

Phoronix Test Suite 9.0 is now available as the latest quarterly feature release to our cross-platform, open-source automated benchmarking framework. With Phoronix Test Suite 9.0 comes a rewritten result viewer to offer more result viewing functionality previously only exposed locally via the command-line or through a Phoromatic Server (or OpenBenchmarking.org when results are uploaded), new offline/enterprise usage improvements, various hardware/software detection enhancements on different platforms, and a variety of other additions.

With the upcoming LLVM Clang 9.0 compiler release there is an amazing achievement more than a decade in the making... The mainline Clang compiler can finally build the mainline Linux x86_64 kernel. The AArch64 state has been in better shape in recent years with multiple Arm vendors using Clang as their default compiler including to build the Linux kernel, but finally in 2019 the mainline Clang can build mainline Linux x86_64. There are a few caveats, but in this article is my experience in doing so with LLVM Clang and the Linux 5.3 kernel as well as running some preliminary benchmarks on AMD and Intel hardware.

With Firefox 69 released and Firefox 70 entering beta, here are some fresh web browser benchmarks between Mozilla Firefox and Google Chrome from Ubuntu Linux. On the Firefox size, Firefox 68, 69, and 70 Beta were tested with and without WebRender being enabled and compared to Google's current Chrome 76 stable release.

Given the recent talk about the Schedutil CPU frequency scaling governor and its future along with CPU frequency scaling behavior in general on AMD Zen 2 processors, here are some benchmarks of the Ryzen 9 3900X when tested with the different Linux "CPUFreq" governor options.

While AMD's hardware folks were launching the EPYC 7002 series, their software crew was pushing out the AMD Optimizing C/C++ Compiler 2.0 with support/optimizations for the Zen 2 micro-architecture. Using the top-end AMD EPYC 7742 in a 2P Linux server configuration, here are C/C++ compiler benchmarks looking at the performance when built by the GNU Compiler Collection (GCC), LLVM Clang, and AOCC 2.0.

Yesterday the SWAPGS vulnerability was made public as a new variant of Spectre V1 that affects all operating systems and is believed to affect only Intel CPUs. The SWAPGS discovery by Bitdefender was quietly mitigated by Microsoft for Windows 10 last month while yesterday the patches were posted for the mainline Linux kernel as the Grand Schemozzle. As soon as learning of this SWAPGS vulnerability and seeing the kernel code, I began running some preliminary performance tests to look at the impact of this latest CPU mitigation.

While this year's GCC 9 compiler release brought initial support for AMD Zen 2 processors with the Znver2 target, the support was sadly incomplete. While the GCC 9 support added some of the new instructions, it wasn't complete (such as RDPRU support remains missing) and the cost tables and scheduler model were not updated from Znver1 to account for the microarchitectural changes. Thankfully, SUSE's compiler experts recently fixed up this support for the GCC 10 compiler and more recently were able to get it back-ported for the upcoming GCC 9.2 for the Linux distributions that will upgrade to that point release. Here are some benchmarks looking at the performance impact of that updated AMD Zen 2 compiler code.

While for Intel x86_64 with the latest compilers it's a very competitive race between LLVM Clang and GCC, how is that battle playing out on the IBM POWER9 front? Using the interesting Raptor Blackbird with IBM POWER9 4-core / 16-thread CPU, here are some recent benchmarks I did between GCC 9, GCC 10, and LLVM Clang 8.

The Linux 5.3 kernel merge window is expected to close today so here is our usual recap of all the changes that made it into the mainline tree over the past two weeks. There is a lot of changes to be excited about from Radeon RX 5700 Navi support to various CPU improvements and ongoing performance work to supporting newer Apple MacBook laptops and Intel Speed Select Technology enablement.