Intel News Archives

Intel today submitted their initial batch of "i915" kernel graphics driver changes to DRM-Next of new driver material slated for the Linux 6.2 cycle.

Intel has published the Intel Extension for TensorFlow that makes use of TF's PluggableDevice mechanism to now provide an Intel GPU back-end for TensorFlow that works with the Data Center GPU Flex Series as well as Arc Graphics discrete GPUs.

While the Intel Core i9 13900K is running fast and well on Linux, a few Raptor Lake IDs have come to light that have been missing from various drivers and only now being addressed.

Merged this morning into Mesa 22.3 for the Intel "Iris" Gallium3D driver is "protected content" support in conjunction with the EGL_EXT_protected_content extension.

While the Intel Alder Lake and now Raptor Lake hybrid processor support on Linux is in good shape after various improvements to the kernel for dealing with the mix of P and E cores, there are occasional caveats. Posted this week were a set of Intel P-State driver fixes around hardware P-states (HWP) calibration to ensure it's working on all Intel hybrid platforms.

A new Linux driver introduced by Intel earlier this year was the In-Field Scan for making use of new silicon failure testing functionality with upcoming Intel server CPUs. The IFS driver and associated hardware capability is for detecting potential problems not caught by parity or ECC checks on systems in production. In-Field Scan was merged in Linux 5.19 but then shortly thereafter the driver was marked "broken" due to some driver design issues coming to light. New patches for IFS have been posted to improve the driver's design and remove that "broken" tag.

In addition to GCC landing patches this week in preparing for Grand Ridge and Sierra Forest, the LLVM/Clang open-source compiler stack has also been seeing patches to prepare for future Intel server processors.

Intel earlier this year more formally announced DAOS as its distributed parallel file-system designed for NVMe storage and aims to be more efficient than other parallel file-systems. Yesterday marked the release of DAOS 2.2 as the newest step forward for Distributed Application Object Storage.

While Intel with the rest of the tech industry continue investing immense resources in areas around AI and talking it up, one of the efforts that has been slow to materialize on the Linux side has been for enabling their Gaussian and Neural Accelerator (GNA) with the mainline Linux kernel. This week the latest Intel GNA driver patches were posted for this neural co-processor.

As part of Intel's work starting to enable the compiler support for their Sierra Forest CPUs, AVX-VNNI-INT8 and AVX-IFMA support has been merged into the GCC 13 compiler for supporting these instructions being first introduced with Intel's Sierra Forest and Grand Ridge.

If as an Ubuntu Linux user you have been held off on purchasing one of the new Arc Graphics discrete graphics cards due to the prospects of having to upgrade your own kernel, Mesa, and linux-firmware packages, Intel has a solution for Ubuntu 22.04 LTS users in the form of a packaged driver.

Last week saw Intel sending out new GCC compiler patches for adding the "Sierra Forest" CPU target and the number of new x86_64 instructions it's adding. Those GCC patches follow Intel publishing an updated programming reference manual where they detailed these new instructions coming for Sierra Forest Xeon CPUs as well as Grand Ridge. Now on the LLVM compiler side, they too have begun landing new patches for these new Intel instructions.

Intel Arc Graphics A750 and A770 work on Linux if you are running the very latest Linux kernel and Mesa. The gaming experience is decent aside from occasional driver issues. One of the games that has been pesky with the open-source Intel "ANV" Vulkan driver has been the HITMAN 3 title running under Steam Play but with the newest Mesa 22.3 code should now be fixed up.

After Intel posted a set of patches last week for the GNU Compiler Collection around Raptor Lake, Meteor Lake, and Sierra Forest, the two more basic patches have already been merged into the GCC 13 code-base while the Sierra Forest Xeon E-core patches and the various new instructions presented there are still undergoing review.

A set of 12 patches adding 720 lines of new code and removing 222 lines of existing code to Intel's "ANV" open-source Vulkan driver have been merged for Mesa 22.3 to help reduce the Vulkan push descriptor CPU overhead.

Last month Intel published the XeSS 1.0 SDK for their Xe Super Sampling technology showcased with Arc Graphics discrete graphics cards. Sadly their initial SDK drop included Windows binaries and wasn't fully open-source. On Friday XeSS 1.0.1 was published with some bug fixes but still not being fully open-source.

Intel has today sent out new compiler patches as they work to get their next-generation processors all set for the open-source GNU Compiler Collection (GCC), including the introduction of some new x86_64 instruction set extensions coming with the high core count Sierra Forest processors.

Intel has published their "2022Q3" releases for their open-source Media Driver with VA-API support as well as their Intel oneVPL GPU Runtime.

Intel on Thursday sent out the initial Linux kernel patches for supporting the LKGS instruction coming with future processors.

In addition to the excitement of the Arc Graphics A750 and A770, Intel has unveiled oneVPL 2022.2 as the newest version of this open-source video processing library that is their flagship video decode/encode library and related video processing toolkit as part of oneAPI.

The Error Detection And Correction (EDAC) driver updates have been submitted and merged for the Linux 6.1 merge window of which there are a few notable additions on the Intel side.

Intel laptop users running Linux are being advised to avoid running the latest Linux 5.19.12 stable kernel point release as it can potentially damage your display.

This week was word of the Intel Arc Graphics A770 launching for $329+ on 12 October, yesterday was the embargo lift on the Arc Graphics A750 also shipping on 12 Ocrober for $289+, and now today is another embargo lift concerning Intel Arc Graphics...

Earlier this week at the Intel Innovation event it was announced the Arc Graphics A770 would be launching 12 October and the base model costing $329 USD. Today the embargo lifts on more details around Intel's forthcoming higher-end Arc Graphics hardware.

The second day of the Intel Innovation event in San Jose featured Intel CTO Greg Lavender talking up the greatness of open standards, open-source, and their wonderful oneAPI initiative. There were a few bits of oneAPI news as part of today's keynote.

Ahead of the flagship Arc Graphics A770 launching on 12 October, Intel's Mesa "ANV" open-source Vulkan driver has finally exposed the ray-tracing support for DG2/Alchemist graphics hardware.

Aligned with Intel's Innovation event happening this week in San Jose, Intel on Tuesday released oneDNN 2.7 as the newest version of their deep neural network library. In addition to optimizing support for new Intel hardware, oneDNN 2.7 also has AMD GPU support.

Yet more news from Intel's Innovation event taking place in San Jose is the initial SDK source code availability of the much anticipated Xe Super Sampling. XeSS is Intel's alternative to AMD FidelityFX Super Resolution and NVIDIA DLSS.

In addition to announcing 13th Gen Core "Raptor Lake" CPUs, Intel also announced their Arc Graphics A770 flagship graphics card will be launching in just two weeks.

Monday saw the AMD Ryzen 7000 series review embargo lift and retail availability beginning today for those Zen 4 desktop CPUs. Intel meanwhile is using today -- the first day of their second annual Innovation event in San Jose -- to announce 13th Gen Core "Raptor Lake" processors for aiming to take AMD's Zen 4 head-on.

Intel at their Innovation conference confirmed that 4th Gen Xeon Scalable "Sapphire Rapids" processors will feature an optional "on demand" activation model.

Intel at their Innovation event this week in California is talking up the new Intel Developer Cloud. The Intel Developer Cloud aims to make it easier for customers/partners/developers to test new and future hardware platforms.

This week saw the Intel Habana Labs AI accelerator driver updates submitted to char/misc ahead of the upcoming Linux 6.1 merge window.

As part of getting their Vulkan ray-tracing support into good shape, a handful of patches were merged today for Mesa 22.3 in fixing up the ANV driver's ray query code.

From two weeks back you may recall the small patches that led to increasing Intel's Vulkan driver draw throughput by ~60%+. Well, as of yesterday the refined version of that work has landed within Mesa 22.3.

While Intel's Compute-Runtime stack is fully open-source and already provides OpenCL 3.0 support for recent generations of Intel graphics under Linux, it looks like the recently-merged "Rusticl" Rust OpenCL implementation in Mesa will soon be working too on Intel graphics hardware as an alternative OpenCL 3.0 implementation.

Intel has become the first corporate "gold" patron to the Krita Foundation for their development fund to advance this open-source digital painting program.

Along with the last drm-intel-gt-next pull for Linux 6.1, a final drm-intel-next pull request of new feature code was submitted to DRM-Next ahead of the upcoming Linux 6.1 merge window.

Adding to the growing examples and early drivers being worked on for the Linux kernel to showcase the possibilities of using the Rust programming language within the kernel, an early port of Intel's e1000 wired networking driver has started.

Back in 2019 is when Intel first detailed the Data Streaming Accelerator "DSA" and began working on the Linux enablement patches. The Intel DSA is designed to help accelerate streaming and transformation operations for storage, networking, and persistent memory. Finally the DSA is coming to market by way of being found within forthcoming Xeon Scalable "Sapphire Rapids" processors while the DSA 2.0 accelerator is already in the works for future processors.

You may recall a few days ago how Valve contractor Mike Blumenkrantz boosted the Radeon Vulkan "RADV" driver draw throughput by +55%. Well, he now had a go at optimizing the Intel open-source "ANV" Vulkan driver and has squeezed out a 60% improvement to the draw throughput. Even more interesting is that it was just a few lines of code.

Intel announced today that beginning with 2023 notebooks, the Intel Pentium and Intel Celeron brands will be replaced by... Intel Processor.

Intel submitted their final set of "drm-intel-gt-next" feature changes intended for merging in the upcoming Linux 6.1 kernel merge window that opens in early October.

For a number of years Arm CPUs on the mainline Linux kernel have supported Energy Aware Scheduling (EAS) as a means of providing the kernel's scheduler with the information to influence its scheduling decisions based on the estimated energy consumed by the CPU cores. EAS employs an energy model for helping to place tasks between the big.LITTLE cores for optimal energy efficiency and a minimal impact on throughput. Intel has been working to eventually support Energy Aware Scheduling on their x86 hybrid CPUs too.

Last month Intel began landing oneVPL support within FFmpeg as their video processing and acceleration library that is part of their oneAPI toolkit. The oneVPL Video Processing Library supports CPU-based execution as well as native Intel GPU acceleration for their latest Gen12/Xe hardware with a focus on Arc Graphics / DG2 hardware, targeting the Intel Media SDK for their older GPUs, and can be adapted for other possible back-ends.

Sent out as a "request for comments" on Friday night was a new patch series out of Intel introducing the notion of "classes of tasks" for the Linux kernel so that the scheduler can make better decisions on hybrid processors like with Intel's Alder Lake processors and upcoming Raptor Lake. This also provides a more complete implementation of Intel Thread Director for Linux and may also be used as the basis for instruction set differences between the performance and efficient cores.

Intel engineers have submitted another batch of "i915" kernel graphics driver changes intended for Linux 6.1 that include updated GuC firmware version handling, more DG2/Alchemist Arc Graphics work, and also more Meteor Lake "MTL" enablement for that successor to Raptor Lake.

In addition to Intel acquiring Linutronix as the company known for their work on the real-time (RT) kernel patches and other contributions and then back in June acquiring Codeplay Software, Intel has today made another notable software talent acquisition... Intel announced this afternoon that the team behind ArrayFire has joined the company to further their ambitious software endeavors.

Intel has detailed more of the Arc Graphics A-Series hardware specifications for upcoming models, including the A700 series.

Blender 3.3 is set to be released today and one of the exciting enhancements with this open-source, cross-platform 3D modeling software update is initial support for Intel oneAPI/SYCL GPU acceleration. Intel Arc Graphics discrete GPUs can now enjoy this accelerated Cycles back-end, permitting your driver stack is new enough on Windows or Linux and are using their new dGPUs and not existing integrated graphics. But this is just the start of their oneAPI GPU-accelerated push for Blender.