Phoronix

NVIDIA engineers have been working on NUMA distance metrics within the Linux kernel to replace the simple local/remote NUMA preference interface currently used by some drivers for NUMA-aware memory allocations. In their testing this improved NUMA distance handling is leading to "significant performance implications" for throughput and CPU utilization...

NVIDIA engineers have been working on NUMA distance metrics within the Linux kernel to replace the simple local/remote NUMA preference interface currently used by some drivers for NUMA-aware memory allocations. In their testing this improved NUMA distance handling is leading to "significant performance implications" for throughput and CPU utilization...

Following the recent discussions about -O3'ing the Linux kernel and other compiler optimizations, a request came in to see some fresh GCC compiler optimization benchmarks with the recently released GCC 12. So here is a fresh look at various GCC optimization levels up through -Ofast as well as with link-time optimizations (LTO) and "-march=native" tuning on the new GCC 12 with the mature AMD Ryzen Threadripper 3990X platform.

Added to Intel's documentation in late 2020 and initial kernel patches out since early 2021, Intel has been slowly working on Linear Address Masking (LAM) support for the Linux kernel. Out this past week was finally the latest iteration of this work for leveraging untranslated address bits of 64-bit linear addresses to be used for storing arbitrary software metadata...

A feature supported by the GNU Compiler Collection (GCC) that sadly isn't used more often is function multi-versioning (FMV) for supporting multiple versions of a function with the given function selected based upon the target processor in use. GCC FMV on x86_64 allows for different functions to be used whether supporting SSE4.2, AVX, or even a particular CPU micro-architecture. Arm is finally working on GCC function multi-versioning support for AArch64...

Nadav Amit who previously spearheaded work on reducing unnecessary TLB flushes, concurrent TLB flushes, and other low level optimizations over the years. The latest work is now on "relaxed" TLB flushes as another low-level performance improvement...

Yesterday marked the release of LightDM 1.32 as the first official release of this display manager since 2019...

In addition to AMD engineers being busy working on RDNA3 graphics support for their open-source Linux graphics driver stack, concurrently they have also been working on enabling GFX940 as their next-gen CDNA part, presumably to launch as the AMD Instinct MI300 if traditions hold...

While Retbleed was grabbing the headlines last week and attention of kernel developers, an Intel engineer sent out a new patch series adding Linux kernel support for "BHI_DIS" as a new hardware-based mitigation that appears to be coming with future Intel CPUs for better fending off Spectre-BHI (Branch History Injection) that was disclosed back in March. But initial indications are this new hardware-based prevention may be even more costly for performance and is not being enabled by default...

Canonical engineer Daniel van Vugt who is known for his work on enhancing the upstream GNOME desktop stack to improve the experience for Ubuntu has recently taken up interest in getting the AMD-Xilinx Kria KR260 working with the GNOME Wayland session. The Kria KR260 is for the recently-announced AMD/Xilinx Robotics Starter Kit...

AMD today published AMDVLK 2022.Q3.1 as their latest snapshot of this open-source Radeon Vulkan driver for Linux systems...

After dealing with the Intel Alder Lake P GuC firmware breakage around Linux 5.19 Git that is now to be addressed by the upstream Intel developers, it was on to performance testing the shiny new Core i7 1280P with this kernel due to be released as stable within the next two weeks... For those concerned about maximum performance, there was a glaring performance regression for this Alder Lake P on the new kernel being released as stable later this month. Well, a default change in performance/behavior at least but the power efficiency / performance-per-Watt tended to be better on this new kernel.

Asahi Linux has issued an update with initial support for the Mac Studio powered by the M1 Ultra SoC as well as having initial (experimental) suppoirt for the new Apple M2 hardware...

Stemming from my article last week noting how Linux 5.19 Git broke Intel Alder Lake P graphics support due to requiring new firmware while not retaining backwards compatibility with the existing Intel GuC firmware, a solution is still being worked on prior to Linux 5.19 final whether it be a revert or the proposed patch working on GuC v69/70 firmware compatibility. Linux firmware guidelines are also being proposed to ensure kernel developers in the future don't try to break firmware support guarantees...

One of several improvements being prepared for the XFS file-system with the upcoming Linux 5.20 cycle is focused on improving the CIL scalability for systems with many CPU cores...

The ASUS-EC-Sensors driver that provides better and faster hardware sensor reading for ASUS motherboards on Linux and premiered earlier this year in Linux 5.18 is continuing to broaden its list of supported ASUS motherboards...

Ben Widawsky who had been at Intel for the past seventeen years, most of which were spent improving their open-source Linux graphics driver as well as other Linux kernel contributions, has joined Google...

Red Hat released Stalld 1.17 this past week as the newest version of this open-source daemon used for thread stall detection and boosting on Linux...

Linus Torvalds just released Linux 5.19-rc7 following a busy week due to the Retbleed security mitigation and not only the CPU overhead performance impact it puts on users but the mess it has on kernel development especially when it comes to embargoed issues that make the patches difficult to review/test well prior to embargo lift...

Following some weekend benchmarks here are more complementary numbers on the Retbleed mitigation performance benchmark costs. These additional numbers are on a Zen 2 based AMD Ryzen 7 4800U APU that has been common both to laptops as well as embedded/low-profile devices for thin client computing, IoT / edge use-cases, and more...