Intel News Archives

Intel today published Compute-Runtime 23.35.27191.9 as their latest update to this open-source GPU compute stack enabling OpenCL and oneAPI Level Zero support on Linux and Windows. With this being their first tagged release since September, it's coming in heavy on changes.

One thing that has never gotten old over the past nearly twenty years of covering Linux news on Phoronix are the relentless performance optimizations made to the Linux kernel, GCC and LLVM/Clang compilers, and other key open-source projects over the years. Intel engineers have been responsible for so many exciting Linux performance optimizations over time at ensuring maximum Linux x86_64 performance as well as ensuring great performance at a macro-level as they've showcased with the likes of Clear Linux. It looks like they have some new innovation(s) in store soon for further maximizing compiler-assisted performance.

Intel Linux kernel graphics driver developers this week sent out their first batch of drm-intel-next i915 DRM driver changes to DRM-Next of new material to be introduced in the upcoming Linux 6.8 cycle.

While much of the modern graphics world these days is focused on the Vulkan API, there's no signs of Intel's open-source graphics driver engineers losing optimization focus with their OpenGL Linux driver by way of the Iris Gallium3D code. Merged this holiday week was a rather significant rework to its buffer object allocation system.

Intel on Thursday committed Arrow Lake "ARL" support to their open-source Intel Graphics Compiler (IGC) that is used by their Compute Runtime stack for OpenCL and Level Zero while on Windows IGC is additionally used as their graphics shader compiler as well.

Intel has announced their 2024 updates to oneAPI that they are promoting as the Intel Software Developer Toolkits 2024 release.

The most significant limitation of Intel Arc Graphics on Linux with the existing open-source driver stack has been the lack of sparse resources support that is needed for many newer games to work via Steam Play on Linux. Intel has a proper solution in place with their yet-to-be-merged Xe kernel driver while now for Mesa 24.0 their ANV Vulkan driver has landed an implementation that works with the existing i915 kernel driver.

Earlier this week Intel compiler engineers posted patches enabling Intel APX NDD support for the GCC compiler as the "New Data Destination" feature of the Advanced Performance Extensions. Those engineers are ending out their week by posting patches for enabling APX PPX for GCC, the new Push-Pop Acceleration of this forthcoming ISA addition.

Intel has released OpenVINO 2023.2 as the newest version of their open-source toolkit for optimizing and deploying AI inference. OpenVINO continues in its never-ending quest for maximizing deep learning performance and efficiency while continuing to expand in its model support and features.

Intel has rolled out a new quarterly update to QATlib, its QuickAssist Technology library for Linux systems to enjoy hardware acceleration of various crypto and compression algorithms on QAT Gen 4 hardware.

Intel has released a new version of ISPC, their Implicit SPMD Program Compiler. The ISPC 1.22 release is clocking in a few percent faster across the board as with their release builds they are now making use of LLVM Clang's link-time optimizations (LTO) for speeding up the binaries.

A new feature coming to next-generation Intel graphics display hardware has been revealed in new open-source Linux graphics driver patches: CMRR as an extension of the existing adaptive-sync variable rate refresh (VRR) functionality.

As part of their work on enabling Advanced Performance Extensions, Intel compiler engineers today posted a set of GNU Compiler Collection patches for enabling the APX NDD.

Intel has released new CPU microcode this morning for mitigating a new CPU security vulnerability (INTEL-SA-00950). This new microcode drop also fixes various functional issues on recent generations of Intel processors.

The Intel-developed Turbostat Linux CLI utility for reporting processor frequency and idle statistics is seeing a number of feature updates for Linux 6.7 as well as new hardware support.

One of the great aspects of Intel integrated and discrete graphics is the broad support for Single Root I/O Virtualization (SR-IOV). Intel "Gen12" graphics back to Tigerlake can handle SR-IOV when there aren't any firmware woes or other issues at play. There is SR-IOV support currently with the i915 kernel driver but Intel engineers are working to architect optimal SR-IOV integration into their forthcoming Xe DRM kernel driver.

Since announcing AVX10 earlier this year, Intel compiler engineers have been quite busy preparing the open-source compiler toolchains like GCC and LLVM/Clang for this next iteration of Advanced Vector Extensions. On Thursday night the latest AVX10.1 work was posted for the GNU Compiler Collection ahead of its upcoming feature freeze.

The Multi-Function Device (MFD) updates were sent out last week for the in-development Linux 6.7 kernel for these drivers catering to heterogeneous hardware blocks.

Last year Intel made available a packaged "Arc Graphics Driver" for Ubuntu 22.04 LTS and later Ubuntu 23.04 to provide a DKMS-backported kernel driver and packaged Mesa driver to make it easier to use Arc Graphics (DG2/Alchemist) during the phase when the upstream kernel support was still stabilizing and not yet found out-of-the-box on Linux distributions at the time. This week marked another rare update for this packaged driver.

More Linux kernel code around Intel's Trust Domain Extensions (TDX) has landed with the in-development Linux 6.7 kernel.

Intel Arc Graphics on Linux can handle running the Diablo IV game via Valve's Steam Play but only when it hides the fact that it's Intel graphics in use via a game-specific driver workaround.

All of the power management related updates have been merged for the in-development Linux 6.7 kernel.

Earlier this year Intel published x86-simd-sort as a very speedy sorting library that initially leveraged AVX-512 instructions for 10x to 17x faster sorts. Numpy was one of the first major projects to adopt x86-simd-sort and OpenJDK more recently adopted it. Since the initial release we've seen more features and performance optimizations added. Today marks the release of x86-simd-sort 4.0 and it's delivering even greater performance while also adding an AVX2 code path to help those without AVX-512.

A shiny feature landed on Friday for the Intel open-source OpenGL and Vulkan drivers within Mesa 24.0: Tile-Based Immediate Mode Rendering (TBIMR).

Intel's FFmpeg Cartwheel is where the company continues to stage their latest FFmpeg multimedia library patches prior to upstreaming. FFmpeg Cartwheel ends up containing all the latest and greatest code for leveraging VA-API and Quick Sync Video (QSV) from Intel integrated graphics through their latest DG2/Alchemist class discrete graphics.

Building off the recent release of OSPRay 3.0 with initial GPU acceleration support, OSPRay Studio has now been updated against this new version. As a reminder, OSPRay Studio is Intel's open-source interactive visualization and ray-tracing application.

While Meteor Lake isn't shipping until December, Intel's open-source Linux engineers for months have already been working on 15th Gen Arrow Lake (as well as Lunar Lake, among their other processor lines) for getting the driver support in order. The latest early hardware enablement to talk about is the initial Arrow Lake S support coming to their video acceleration drivers.

Following an FCV optimization for the latest Intel graphics hardware, ASTC LDR emulation, some still-pending Vulkan sparse support for ANV atop the existing i915 driver, and other recent Intel open-source "ANV" Vulkan driver optimizations, another optimization was just merged into Mesa 23.3.

An Intel engineer on Friday posted a set of Linux kernel patches that are working to refine the Microarchitectural Data Sampling (MDS) mitigation handling for the Linux kernel to better protect some kernel data and also some very subtle performance benefits.

With the pending changes set to be merged for the upcoming Linux 6.7 kernel, all known issues/limitations around the i915 kernel driver support for upcoming integrated graphics with Intel Meteor Lake laptop processors.

Following other Intel oneAPI components like Embree and OpenVKL introducing GPU acceleration via SYCL, today Intel released the open-source OSPRay 3.0 that rolls out initial GPU support for this portable ray-tracing engine.

In addition to Intel putting the finishing touches on Meteor Lake graphics support for the upcoming Linux 6.7 cycle, Intel's Habana Labs AI accelerator driver also has some notable changes for this next kernel version.

Intel continues to do a splendid job at ensuring the open-source GCC and LLVM/Clang upstream compilers have support for their new processor cores well in advance of products shipping. Beyond already having Sierra Forest, Granite Rapids, Lunar Lake and Arrow Lake support already in upstream GCC, today one of the Intel compiler engineers sent out patches for enabling the Clear Water Forest and Panther Lake cores.

The USB I2C / SPI / GPIO drivers for Intel's La Jolla Cove Adapter (LCJA) are set to be mainlined with the upcoming Linux 6.7 kernel merge window opening at the end of the month.

Next week in DRM-Next will hopefully see the patch promoting Intel Meteor Lake graphics to stable in time for the upcoming Linux 6.7 merge window. Ahead of that an i915 drm-intel-next pull request on Friday sent out more Meteor Lake patches while concurrently working on more driver enablement code for Lunar Lake.

Similar to the Radeon RADV driver recently implementing software-based decoding for Adaptive Scalable Texture Compression (ASTC), the Intel "ANV" Vulkan driver within Mesa has also now wired up ASTC LDR emulation.

It was a busy day for the GCC 14 (GNU Compiler Collection) development with Intel landing three separate noteworthy contributions just today for this next major open-source compiler release.

While one month ago a Linux kernel patch was floated for advertising Meteor Lake graphics support by default to effectively mark it as stable and remove it from behind the "i915.force_probe" block, that has yet to be queued for the mainline kernel. Today another set of patches were submitted of new Intel kernel graphics driver changes slated for Linux 6.7 with this patch still missing -- but it might squeeze in next week to still make it for the v6.7 cycle.

Released back in May was Intel's open-source Open Image Denoise 2.0 that brought GPU support via SYCL for this denoising library intended for use with ray-tracing applications. Following that big release as part of the broader effort to make Intel's oneAPI suite more supportive on GPUs/accelerators, OIDn 2.1 released on Wednesday with fixes and performance improvements to the GPU support.

Intel has enabled a fast clear optimization "FCV" for their Gen12.5 graphics and newer under Linux with the open-source "ANV" Vulkan driver. This can help games like F1 22 with improving performance as much as 45%.

Intel submitted a new i915 DRM Linux kernel driver patch for adding four additional device PCI IDs for DG2/Alchemist graphics processors.

Intel has formally announced the Arc Graphics A580 as their latest graphics processor to fit between the low-end A380 and higher-end A750/A770 graphics cards.

In addition to Intel's compiler engineers pushing a lot of code into GCC -- and other open-source compiler components -- around AVX10, over the weekend code began hitting the GCC 14 Git codebase for the Advanced Performance Extensions (APX) functionality.

In addition to x86-simd-sort 3.0 being released for speedy AVX-512 sorting, Friday also brought the release of oneDNN 3.3 as the deep neural network library that is part of oneAPI and focused on helping developers build out deep learning applications.

Earlier this year Intel posted x86-simd-sort as a blazing fast sorting library that makes use of AVX-512. When the popular Numpy began using it they found up to 10~17x faster sorts for 16-bit to 64-bit data types. Today Intel software engineers released x86-simd-sort 3.0 and it also comes minutes after OpenJDK merged a modified version of this speeding sorting code into that reference JDK codebase.

Intel Linux graphics driver engineers are doing a bit of house keeping to the i915 DRM driver and removing some pre-production hardware bits as well as the incomplete Ponte Vecchio support.

In addition to Vulkan sparse support that works with the existing i915 kernel driver, another exciting open-source Intel Vulkan driver development this week is landing a transfer queue implementation for DG2/Alchemist GPUs. This in turn can help with hybrid graphics systems and other situations.

Motivated by a 25% performance degradation seen on an Intel Xeon Scalable dual socket server due to Indirect Branch Restricted Speculation (IBRS), Red Hat's Waiman Long has been working on a patch series to update the IBRS handling in different conditions for affected Intel processors on Linux.

The biggest hindrance for using Intel Arc Graphics for Linux gaming has been the lack of Vulkan sparse support as needed for running many newer Windows DirectX 12 games atop Valve's Steam Play with Proton using VKD3D-Proton. Intel recently did implement Vulkan sparse support for ANV in Mesa 23.3 but it only works with their yet-to-be-upstreamed and still-experimental Xe kernel driver. But now Intel Linux graphics driver engineers have managed to pull off a solution for getting the sparse resources supported while using the existing i915 kernel DRM driver.

Similar to Embree 4.0 adding Intel graphics acceleration earlier this year via SYCL, Intel's OpenVKL library that is also part of their oneAPI rendering toolkit has now added SYCL graphics acceleration with OpenVKL 2.0.