Linux 5.16 Features Include FUTEX2, Intel AMX, Folios, DG2/Alchemist, More Apple Silicon Support

Written by Michael Larabel in Software on 15 November 2021 at 09:10 AM EST. Page 2 of 2. 8 Comments.

Linux Storage / File-Systems:

- Block subsystem optimizations including much of Jens Axboe's work on optimizing the per-core IOPS potential of the Linux kernel.

- More performance improvements for Btrfs.

- F2FS adds an option to intentionally fragment the file-system for developer benefit.

- Faster Ceph with async dirops enabled by default.

- AFS, 9p, and Netfslib now make use of folios. Memory folios were merged this cycle and those initial users are ready while XFS migrating to folios and the like is not expected until at least Linux 5.17.

- LZMA/MicroLZMA compression for EROFS.

- Memory footprint reduction work for XFS.

Networking:

- SMB3/CIFS improvements fron Microsoft including fixes and some performance work.

- Realtek RT89 WiFi driver for supporting new 802.11ax wireless adapters.

- The usual plethora of networking subsystem updates.

Other Hardware:

- Working sensor support for many more ASUS and ASRock motherboards.

- Support for the 2021 Apple Magic Keyboard.

- The Habana Labs AI driver now supports peer-to-peer sharing via DMA-BUF.

- The ACPI-side work for allowing a driver to probe hardware while powered off or in a low-power state.

- More CXL subsystem enablement work.

- System76 laptop hardware support improvements.

- A new driver for dealing with EC-driven backlights.

- Better AMD S0ix support.

- USB work as part of the Apple Silicon bring-up.

- The Apple M1 PCIe driver was merged as another big step forward to Linux support for Apple Silicon.

- AMD Yellow Carp run-time power management for the XHCI controllers.

- Many different power management improvements.

- Better USB low-latency audio support and other sound improvements.

Security:

- SELinux / LSM / Smack controls and auditing for IO_uring.

- Improved Retpoline code for dealing with how the return trampoline code is rewritten. The x86 BPF code also now aligns better with expectations around Retpolines.

- Preparation work for supporting FGKASLR in the future as fine-grained / function granular kernel address space layout randomization.

- Support for KVM guests to have control over the AMD PSF control bit for making that security-related change if desired. The Linux kernel itself still has not landed that specific AMD Predictive Store Forwarding control option.

- Microsoft began landing Hyper-V isolation VM support.

- Loosening of the Spectre SSBD / STIBP defaults for SECCOMP threads.

Other:

- Memory folios has landed as a core improvement to the Linux memory management code.

- DAMON-based memory reclamation is merged for helping Linux in low-memory situations.

- The updated Zstd implementation for the kernel has finally made it! After years of falling out-of-date, the new implementation should be easier to maintain/update from upstream more easily. The big update to the Zstd kernel code should yield significant compression/decompression performance improvements for the kernel users from compressed modules to Btrfs and F2FS Zstd-based native file-system compression.

- Xen can handle faster booting of PV guests.

- Lots of cleaning to the staging code.

Onward now to more Linux 5.16 kernel benchmarking.

If you enjoyed this article consider joining Phoronix Premium to view this site ad-free, multi-page articles on a single page, and other benefits. PayPal or Stripe tips are also graciously accepted. Thanks for your support.


Related Articles
About The Author
Michael Larabel

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.