Software Linux Reviews & Articles

There have been 951 Linux hardware reviews and benchmark articles on Phoronix for software. Separately, check out our news section for related product news.

Stemming from a request by a Phoronix Premium reader wondering about some fresh historical kernel performance comparison numbers, today's benchmarking is looking at the performance of the LTS and latest stable Linux kernel versions going back to Linux 5.15 LTS in 2021. For testing an AMD EPYC Milan-X server was used for compatibility back through Linux 5.15 LTS with some rather impressive results for testing these major Linux kernel releases of the past four years.

With the Linux 6.17-rc1 release this past Sunday, the Linux 6.17 merge window is over. Here is a look back at the most exciting changes that made it for Linux 6.17.

With the newly-released Linux 6.16 kernel there is the new X86_NATIVE_CPU build option if wanting to optimize your kernel build for your local CPU in use. Enabling CONFIG_X86_NATIVE_CPU is setting the "-march=native" compiler optimizations for the kernel build in an effort to ensure peak performance/optimizations for the local system. Here are some benchmarks looking at the impact of X86_NATIVE_CPU on Linux 6.16 while using the HP ZBook Ultra G1a laptop with AMD Strix Halo SoC as an interesting test target for squeezing additional performance.

For those on an Intel Core Ultra Series 2 "Lunar Lake" system, the upcoming Linux 6.16 kernel is looking to be in better shape for those newest Intel SoCs. In testing carried out using a Lenovo ThinkPad X1 Carbon Gen 13 Aura edition laptop, there are performance gains in some areas with the Linux 6.16 development kernel.

For those curious about the direction of Mozilla Firefox web browser performance over the past year and a half, here are web browser benchmarks for every Firefox release from Firefox 120 in November 2023 through the newest Firefox 140 stable and Firefox 140 beta releases from a few days ago. Every major Firefox release was benchmarked on the same Ubuntu Linux system with AMD Ryzen 9 9950X for evaluating the performance and memory usage of this open-source web browser.

Following this week's release of Firefox 140, Firefox 141 was promoted to beta. Most exciting for Linux users with next month's Firefox 141 release is finally lowering system RAM use! I've been running some benchmarks looking at the impact.

While the AMD Ryzen AI Max+ PRO 395 and Ryzen AI Max+ 395 are very similar processors just as the Ryzen PRO processors are to other non-PRO parts, one of the differences with the AMD PRO Technologies come down to AMD Memory Guard providing full system memory encryption. From the HP BIOS with the ZBook Ultra G1a there is a convenient toggle for this full memory encryption support and thus I decided to carry out some benchmarks to measure the performance cost to this memory encryption feature on this AMD Strix Halo SoC.

The Linux 6.15 kernel cycle started off a bit rough with a heavy hitting performance regression spotted and then fixed but to only then discover another Linux 6.15 performance regression affecting modern AMD CPUs. Fortunately those issues were cleared out in time for the recent Linux 6.15 stable release. Linux 6.15 stable is looking good especially on 5th Gen AMD EPYC "Turin" servers with some recent benchmarks showing some modest gains over the Linux 6.14 kernel.

Now that the Linux 6.16 kernel merge window closed this weekend, here is a convenient recap of all the interesting features coming in this next kernel release due out as stable around the end of July.

Since recently picking up a Qualcomm Snapdragon X Elite laptop for Linux testing now that the Linux support is starting to evolve into better shape (albeit still with many rough corners limiting the daily usability of such devices with Linux), one of the areas I was curious about was looking at the performance of Linux binaries with GCC vs. LLVM Clang. Here are some benchmarks for those wondering how the GCC and Clang compilers are competing on the Snapdragon X Elite with the Oryon CPU cores.

With the AMD Ryzen Threadripper 9000 series having just been announced at Computex, it's a good time to revisit the Ryzen Threadripper 7000 series performance ahead of those Zen 5 HEDT CPUs launching in July. In this article is a look at how the AMD Ryzen Threadripper 7970X 32-core and 7980X 64-core HEDT processors have evolved on the same system with the software updates released since their late 2023 debut. Overall the AMD Ryzen Threadripper 7970X and 7980X Linux performance is up by about 8% overall since launch day less than two years ago with the 64-core processor.

Given the significant gains to the Intel Arrow Lake performance on Linux since launch that I showcased last month on Phoronix, I was also eager to see how the Intel Lunar Lake performance has evolved since its debut last year. In this article is a look at those relatively weak launch-day Linux performance numbers for the Core Ultra 7 256V compared to where they are performing now on Ubuntu 25.04 for showing how far the Intel Lunar Lake performance has evolved since last September.

With the copy-on-write Bcachefs file-system considering its on-disk format now "soft frozen" and nearing the point of potentially removing the "EXPERIMENTAL" flag on it, a number of Phoronix readers have been requesting some fresh benchmarks of this open-source file-system. For your viewing pleasure today are some fresh benchmarks of Bcachefs and other file-systems atop the Linux 6.15 kernel being released as stable later this month. On the benchmarking block today are Bcachefs, Btrfs, EXT4, F2FS, and XFS in-tree file-systems.

With the spring Linux distribution/OS updates upon us, in recent weeks I've looked at the Ubuntu 25.04 performance gains on AMD EPYC 9005 "Turin" and also the performance of Fedora Server 42. Following that I expanded the scope of the Linux operating systems (distributions) benchmarks on the latest 5th Gen AMD EPYC server hardware. Here is a look at how the performance of the new Ubuntu and Fedora Linux releases compare to AlmaLinux and Intel's in-house Clear Linux distribution that tends to be at the forefront of open-source performance optimizations.

With the recently released GCC 15 (GCC 15.1) compiler besides adding new language features, enhancements to help developers in debugging build failures, and other refinements, there is the never-ending quest of compiler performance optimizations. Since the recent GCC 15.1 release candidate I've been testing this annual compiler feature release on more hardware, including several AMD 5th Gen EPYC "Turin" servers to great success compared to the prior GCC 14 stable series.

Separate from last week in uncovering a big performance regression on Linux 6.15 affecting workloads like Nginx and that regression getting fixed, I unfortunately discovered another heavy-hitting regression on Linux 6.15. This latest performance regression has been bisected and a possible fix is being thought through by the relevant party, but for the moment has yet to be fixed upstream and affects modern AMD processors.

The Linux 6.15 kernel has just merged a fix for the big performance regression I spotlighted yesterday on Phoronix with a huge hit to the Nginx HTTPS web server performance that could see a 3x regression from the in-development Linux 6.15 kernel code. It turns out other workloads/applications also were negatively impacted by this regression. While a stumper at first even with the bisected commit, the issue was luckily resolved very quickly.

With the Linux 6.15 kernel settling down nicely, I've been testing out the current Linux Git state on more systems in looking for any performance changes. Unfortunately this week I ran into a large performance regression affecting the Nginx HTTP(S) web server. Here's a look at that problem currently affecting Linux Git.

The Linux 6.15 merge window ended on Sunday with the release of Linux 6.15-rc1. There is a lot of exciting features and updates that were merged during the two-week merge window. Here is a look at all of the most prominent changes to be found with Linux 6.15.

Last month I posted benchmarks showing the performance when using the new 3D V-Cache Optimizer driver on Linux using the flagship Ryzen 9 9950X3D. This optimizer driver allows tuning the "amd_x3d_mode" for indicating your preference for the CCD with the higher frequency or larger cache size. For some additional insight into the 3D V-Cache Optimizer driver performance impact on Linux, here are benchmarks looking at the difference while using the AMD Ryzen 9 9900X3D.

Last week I published an article looking at the power efficiency of 5th Gen AMD EPYC "Turin" using the modern AMD P-State driver and the Power Profile options. The AMD P-State driver being used by default now for the EPYC 9005 series processors on Linux 6.12+ and paired with Power Profile option tuning can deliver a nice boost to server energy efficiency with only modest impact to the performance. Today's article are some complementary numbers I carried out recently on a Supermicro server looking at more of the Power Profile Selection options.

Last week NeatLab released Neat Video 6 as the newest version of their video engine to reduce noise and enhance video quality that can be used with the likes of DaVinci Resolve, Final Cut, and Premiere Pro software. Neat Video 6 features a faster render engine, optimized CPU + GPU performance, enhanced memory management, and other improvements. Neat Video 6 continues to support Linux natively when it comes to DaVinci Resolve usage as well as various OFX hosts like Natron / Flame / Mistika / Fusion Studio / Nuke. As they also updated their NeatBench benchmark, I was curious to see how the performance of Neat Video 6 is for the new NVIDIA GeForce RTX 50 Blackwell GPUs.

In the recent discussion over the GNU Gold linker being deprecated, there was the usual LLVM vs. GCC compiler/toolchain debate. Fortunately, with recently working on some initial benchmarks of the GCC 15 compiler I was following that up with some fresh LLVM Clang compiler comparison metrics on the same AMD Zen 5 hardware.

With the GCC 15 compiler having progressed to its final stage of development prior to the GCC 15.1 stable release in the likely March~April time frame, I've begun testing the updated GNU Compiler Collection on some test systems. Overall GCC 15 is looking nice and on AMD Zen 5 "znver5" in particular seeing some solid gains over GCC 14. Here are some initial performance benchmarks of the GCC 15 compiler.

Now that the Linux 6.14 merge window wrapped up this past weekend with the release of Linux 6.14-rc1, here is a recap of all the great new features, hardware enablement, and other improvements to find with this kernel.

With Linux 6.14-rc1 released I have begun trying out the new development kernel on a few systems locally. At least for high core count hardware tested thus far, Linux 6.14 at the moment during this early testing phase is sporting some performance regressions within some multi-threaded workloads.

Last weekend a Meta engineer posted Linux kernel patches to make use of the AMD INVLPGB instruction for broadcast TLB invalidation. The Linux kernel can in turn invalidate TLB entries on remote CPUs without needing to send IPIs and without having to wait for remote CPUs to handle those interrupts. Synthetic benchmarks shown in that patch series were very promising and thus I carried out some benchmarking over the holidays of this AMD INVLPGB support for the Linux kernel.

Linux 6.12 was recently promoted to being this year's Long Term Support (LTS) kernel with it being the last major kernel release of 2024. For those enterprise Linux users, hyperscalers, and others typically jumping from one annual LTS kernel to the next, in this holiday article are some benchmarks looking at the performance benefits of Linux 6.12 LTS compared to Linux 6.6 LTS while testing on an AMD Ryzen Threadripper workstation.

This past week Intel published an Intel Core Ultra 200S Series "Arrow Lake" performance status update following mixed reviews since launch around the Arrow Lake gaming performance that were inconsistent with Intel's internal findings. Among Intel's findings detailed in their report this past week were some new BIOS performance optimizations, some misconfigured performance settings in early/reviewer BIOSes, and also some Windows 11 updates being pushed down to help with different performance issues. ASUS already started releasing new BIOSes that incorporate the 0x114 Arrow Lake intended to help the situation. While it's been a Windows-focused issue, I couldn't help but to run Intel Arrow Lake performance comparison benchmarks on Linux with the new microcode / BIOS.

Last month I looked at the impact of AMD's BIOS tuning guide on AI / machine learning workloads for new 5th Gen EPYC "Turin" processors. In today's article I am looking at the performance and power efficiency impact of AMD EPYC 9005 series processors with AMD's BIOS tuning recommendations for Java workloads on Linux.