Phoronix

A batch of fixes to the AMDGPU kernel graphics driver were sent in today for Linux 5.9. While AMDGPU fixes this late in the kernel cycle tend to not be too notable, this time around there are some prominent items worth covering...

Following last week's news of Firefox Nightly flipping on their new JIT "Warp" update I was eager to run fresh benchmarks of the current Firefox releases compared to Google Chrome under Ubuntu Linux.

Just in time for the end of the quarter Intel's open-source multimedia team has released the Media Driver 2020.3 package for the Intel graphics accelerated media encode/decode component on Linux platforms...

Back in April was the provisional release of OpenCL 3.0 with making CL 2.x features optional while adding async DMA extensions and more. Today the finalized version of OpenCL 3.0 has been released plus also introducing an official Khronos OpenCL SDK...

The sixth spin of Intel DG1 discrete graphics card patches have now been sent out for review, amounting to just about 700 lines of new driver code due to building off the existing DG1 work and more broadly the Gen12/Xe support that's been refined in mainline for months. With these patches it would appear the Intel DG1 is then in good shape under Linux but due to the timing is unlikely to be mainlined until a stable kernel release in early 2021...

Last week on the GNU toolchain side was initial work on supporting Intel Key Locker while this week Key Locker support has come to LLVM...

NVIDIA's Linux Vulkan beta driver build has moved from the 450 series that it's been on for a while to the current 455 branch...

NVIDIA software engineer Zi Yan sent out on Monday his latest "1GB PUD THP" patches in aiming to boost application performance on Linux for software making use of large amounts of RAM...

For those making use of the "MT76" WiFi driver for Mediatek MT76 series wireless support, the Linux 5.10 kernel should be a nice upgrade...

As we approach the end of Q3'2020 there have been 783 original news articles on Phoronix this quarter and another 40 featured reviews / multi-page articles. Here is a look back at what's been keeping readers informed during this turbulent year...

Intel's compiler experts in Moscow have been working on OpenCL C 3.0 support for the LLVM Clang compiler front-end...

The open-source Panfrost graphics driver, which is now backed/supported by Arm after starting as a reverse-engineering effort, has picked up support for the Mali G72 GPU...

In addition to Fedora Workstation 33 switching to Btrfs, there are a number of key components updated in Fedora 33 as well as finally enabling link-time optimizations (LTO) for package builds that make this next Fedora Linux installment quite interesting from a performance perspective. Here are some initial benchmarks of Fedora Workstation 32 against the Fedora Workstation 33 Beta on an Intel Core i9 10900K system.

The beta of Fedora 33 is available this morning ahead of the official release expected at the end of October for this Red Hat sponsored Linux distribution...

While yesterday brought the release of Intel's oneAPI 1.0 specification, the interesting news today is that oneAPI support is coming to AMD Radeon graphics cards...

Following the news yesterday of NetBSD changing its default X11 window manager after two decades with TWM to now using CTWM by default, some wondered why they don't jump on the Wayland bandwagon...

MoltenVK 1.1 is out as a big update for this graphics translation layer for getting the Vulkan API running on macOS and iOS devices by translating calls to Apple's Metal API...

Mesa 20.2 has managed to release just before the end of the the quarter. This Mesa Q3'2020 graphics driver update is coming out about one month behind schedule but the wait is worthwhile given many open-source OpenGL and Vulkan driver updates...

Mir 2.1 has been released as Canonical's project around offering a set of libraries for constructing Wayland shells particularly with Snap confinement support and other Ubuntu-focused features...

Hobbyist developers are trying once again to get a Motorola 6800 back-end merged into the upstream LLVM compiler. Yes, the M68k processors that are some 30+ years old...