Graphics: NVIDIA, AMD and Intel's Latest

• NVIDIA release the 460.39 Linux driver update, improved support for kernel 5.10+

  Ready for a new stable NVIDIA driver for Linux? There's a new one out now with added GPU support and some tidying up work done with bug fixes too.

• AMD RDNA2 "Duty Cycle Scaling" Will Turn Off The GPU Under Heavy Load For Relief

  A new Radeon power management feature with RDNA2 graphics processors being exposed by the open-source Linux driver is Duty Cycle Scaling in the name of power/thermal management with a focus on low-power hardware. AMD graphics Duty Cycle Scaling is designed for "small power limit SKUs" and is designed to actually shut off the graphics core and power it back up based on current/power/temperature thresholds. Under heavy workloads, the AMD "DCS" functionality controlled by the graphics firmware will power down the GPU during heavy load scenarios for power/thermal relief before being powered back up to resume work.

• Linux 5.12 Bringing VRR / Adaptive-Sync For Intel TIger Lake / Xe Graphics - Phoronix

  Finally with the upcoming Linux 5.12 cycle is support for Variable Rate Refresh (VRR) / Adaptive-Sync for Intel Tiger Lake "Gen12" Xe Graphics and newer.  The VRR/Adaptive-Sync support for the latest-generation Intel graphics with Tiger Lake and the likes of the forthcoming Rocket Lake, Alder Lake, and discrete DG1 graphics is now in order for the mainline kernel. The VRR enabling for Tiger Lake and newer was sent in this morning to DRM-Next ahead of the Linux 5.12 kernel. This effort has been going on for many months while now has reached the stage that it's ready for merging. 

• Intel Iris Xe Discrete Card Will Only Work With Select CPUs and Motherboards

  These motherboards require a special BIOS that supports Intel Iris Xe, so the cards won’t be compatible with other systems.

Linux 5.10 LTS Status and More Stables Releases of Linux Announced Today

• Linux 5.10 LTS Will Only Be Maintained Until EOY 2022 Unless More Companies Step Up

  Announced a few years ago was the notion of "extended" LTS kernel versions whereby the long term support cycle would span six years rather than the usual two years for LTS kernels in providing maintenance and bug/security fixes to the codebase. This means Linux 5.4 LTS is supported until the end of 2025, Linux 4.19 until the end of 2024, and even Linux 4.14 until the start of 2024. But with the recently minted Linux 5.10 LTS at least for now it's only being committed to maintenance until the end of next year. There's been differing remarks/indications for how long the Linux 5.10 long-term support cycle would last with many expecting six years given that's what has been happening on recent LTS kernels -- even the Linux 4.4 kernel is being planned for maintenance until February 2022 and Linux 4.9 until 2023. Linux stable maintainer Greg Kroah-Hartman has now provided a more transparent answer on the Linux kernel mailing list stemming from the talk over how long Linux 5.10 will be maintained.

• Three stable kernels

  Stable kernels 5.10.11, 5.4.93, and 4.19.171 have been released. They contain important fixes and users should upgrade.

• 5.10.11

• 5.4.93

• 4.19.171

Security and FUD

• Security updates for Wednesday

  Security updates have been issued by Arch Linux (sudo), CentOS (sudo), Debian (sudo), Fedora (kernel, php-pear, and sudo), Gentoo (cacti, mutt, and sudo), Mageia (sudo), openSUSE (sudo), Oracle (sudo), Red Hat (sudo), Scientific Linux (sudo), Slackware (sudo), SUSE (go1.14, go1.15, nodejs8, and sudo), and Ubuntu (libsndfile and sudo).

• Mimecast admits certificate compromise tied to SolarWinds supply chain attack

  Email security firm Mimecast has admitted that the compromise of a certificate it had issued for some Microsoft services is connected to the SolarWinds supply chain incident.

• SolarWinds Cyberattack: Layered OT Security Creates Best Defense

  The recent SolarWinds supply chain cyberattacks serve to underscore an age-old cybersecurity tenant, and the reason we need to continue beating the drum as cybersecurity professionals: Use a layered approach to OT security. This incident highlights a rare, specific use case of a nation state threat actor, an Advanced Persistent Threat (APT). In this particular case, layers provided somewhat limited value, but helped keep the less skilled attackers – about 99% of those on the playing field – at bay. Technology boundaries can be used to lessen the impact of (but unfortunately not prevent) nation state APTs. They not only offer additional protection, they may also help expose the presence of APTs in your network. Let’s examine how they would have helped in the case of APTs like the Sunburst malware that infected SolarWinds Orion software and was downloaded by 18,000 organizations.

• Linux malware uses open-source tool to evade detection [Ed: How pro-Microsoft propaganda sites blame for a tool which comes from Microsoft (GitHub) "Open Source" and "Linux" (though it is the fault of neither). Alternative headline: Microsoft malware is being used to attack machines that run GNU/Linux]

  This tool is known as libprocesshider and is an open-source tool available on Github that can be used to hide any Linux process with the help of the ld preloader.

AMD Schedutil vs. Performance Governor Benchmarks On Linux 5.11 Shows More Upside Potential

With a pending patch, the Linux 5.11 AMD Zen 2 / Zen 3 performance is looking very good as far as the out-of-the-box performance is concerned when using Schedutil as is becoming the increasingly default CPU frequency scaling governor on more distributions / default kernels. With the previously noted Linux 5.11 regression addressed from when the AMD CPU frequency invariance support was first introduced, the Schedutil performance from small Ryzen systems up through big EPYC hardware is looking quite good. But how much upside is left in relation to the optimal CPU frequency scaling performance with the "performance" governor? Here is a look at those benchmarks on Ryzen and EPYC for Schedutil vs. Performance on a patched Linux 5.11 kernel.