Upstream developers are looking at phasing out Intel MMX that was popular in the late 90's but has since long been succeeded by SSE and AVX instruction set extensions.

Due to potentially problematic/incorrect usage around MMX intrinsics usage if the programmer doesn't correctly handle the intrinsics and there being a number of LLVM bugs about MMX behavior, a proposal was issued this past weekend to remove MMX support.

The initial proposal was to re-implement all the MMX intrinsics in Clang's headers by using SSE/SSE

SHADERed has been available for a while now as a cross-platform, open-source shader editor and for debugging and writing of graphics shaders. SHADERed not only runs on Linux and Windows now but through any modern web browser as of the new v1.4 release.

SHADERed 1.4 released today and comes with a web browser port for those wanting to work on shaders within the confines of your web browser.

SHADERed supports working on vertex/pixel shaders as well as the likes of compute and geometry shaders.

It's not clear if AMD has provided the independent RADV Vulkan driver developers at the likes of Valve, Red Hat, and Google with any Navi 2 hardware yet, but they do seem to be making progress on this open-source Radeon Vulkan driver separate from their official AMDVLK open-source driver.

Over the course of summer there was initial RADV support for Sienna Cichlid followed by Navy Flounder. Both these parts are unreleased GFX10.3 Radeon GPUs that appear to most definitely be Navi 2 at this point

Even more optimizations have now been queued up for GNOME 3.38 launching this month.

On top of a number of performance optimizations that landed over the past half-year, hitting Mutter's code-base this weekend for the imminent 3.37.92 milestone is another tweak. Canonical's Daniel van Vugt provided an improvement to cull actors that don't intersect the redraw clip.

This optimization is quite beneficial as he explained in the merge request , " This was inspired by the activities

While LLVM's Clang C/C++ compiler was traditionally known for its faster build speeds than GCC, in recent releases of GCC the build speeds have improved and in some areas LLVM/Clang has slowed down with further optimization passes and other work added to its growing code-base. As it stands right now, GCC is faster than Clang at compiling the Linux kernel.

Presented at last week's Linux Plumbers Conference 2020 was a look at the kernel compile times with Clang. What developers

Just over twenty years after the Linux From Scratch project was started as a guide/book to building all of the software components manually from source, Linux From Scratch 10.0 has been released.

Linux From Scratch continues to teach the user how to roll their own Linux installation from source and while hearing "LFS" has been less common in recent years the effort still continues.

With Linux From Scratch 10.0, the book has gone through a "major reorganization" to cover more cross-compilation techniques

Earlier this year Amazon announced Bottlerocket as a Linux distribution for running containers . This week Bottlerocket crossed the general availability milestone.

Bottlerocket from Amazon Web Services is a Linux distribution optimized for running containers with an emphasis on security, operations, and manageability. Bottlerocket is open-source and available via GitHub . A lot of Bottlerocket's own tooling is written using the Rust programming language.

Besides highly leveraging the Rust programming language, Bottlerocket makes use of eBPF, the dm-verity target, SELinux, and other recent innovations around

Intel's open-source engineers today sent out their latest patches bringing up the Dynamic Load Balancer 2.0 for the next-gen PCIe device that offers load-balanced, prioritized scheduling of core-to-core communication.

DLB 2.0 is the next-generation version of the Intel Dynamic Load Balancer. This is a PCIe device part of the CPU package for providing load balancing features and can take responsibility for distributing system workloads for maximizing the performance -- "significantly greater performance" than software load-

One important bit not covered in today's GeForce RTX 3070/3080/3090 announcement but now detailed via the NVIDIA website is confirmation that the RTX 30 "Ampere" GPUs do in fact have dedicated AV1 hardware decode capabilities.

The RTX 30 series is NVIDIA's first line of GPUs supporting AV1 decode (though sadly no AV1 encode) for this open-source, royalty-free video codec that competes with H.264/