Earlier this month I wrote about Intel SST Core-Power patches as part of Intel's Speed Select's functionality for more control over per-core power/frequency behavior based upon the software running on each core. The "core-power" profile support appears ready now for Linux 5.6.
Intel News Archives
2,934 Intel open-source and Linux related news articles on Phoronix since 2006.
Patches written two months ago for Intel's ANV open-source Vulkan driver have now been merged ahead of the imminent Mesa 20.0 feature freeze and branching.
As more last minute work for the upcoming Mesa 20.0 is initial OpenGL tessellation support for Intel's OpenSWR driver.
After missing their original target of transitioning to Intel Gallium3D by default for Mesa 19.3 as the preferred OpenGL Linux driver on Intel graphics hardware, this milestone has now been reached for Mesa 20.0!
Last year Linux kernel developers began removing support for Intel's Memory Protection Extensions (MPX) and that looks like it will be finished up in the forthcoming Linux 5.6 cycle.
The Khronos SYCL standard as a single-source C++-based programming model for OpenCL is one of the exciting elements for Intel's GPU compute plans with the forthcoming Xe graphics cards and fits into their oneAPI umbrella. They just released their SYCL Compiler and Runtimes 2019-12 release with numerous updates.
For months we have seen various Intel open-source Linux graphics driver patches that begin preparing for multi-GPU support where in moving forward with their Xe graphics cards there could be the iGPU + dGPU setup or even multiple Xe graphics cards in a single system. So far those Intel Linux multi-GPU preparations have been focused on their kernel-space driver while now it's reaching into user-space with their Vulkan driver seeing early infrastructure changes.
An area where Intel continues striking with rhythm and near perfection is on the open-source software front with their countless speedy and useful open-source innovations that often go unmatched as well as timely hardware support. Out this weekend is their OSPray 2.0 release for this damn impressive ray-tracing engine.
As for CVE-2019-14615 the Intel graphics vulnerability disclosed this week affecting Gen7 through Gen9 graphics architectures, it's been dubbed "iGPU Leak" by the researchers involved. Thanks to the researcher who originally discovered this vulnerability having reached out to us, we now have some more information on this issue they describe as a "dangerous vulnerability."
We haven't heard of the Simple Firmware Interface in a number of years, but that changed this week in Linux now formally marking SFI as "obsolete" and confirmation Intel does not plan to ship any future platforms with this standard that dates back to their early days of working on Atom-powered mobile devices.
On Wednesday I shined the light on the initial performance hit from Intel's CVE-2019-14615 graphics vulnerability particularly striking their "Gen7" graphics hard. That initial testing was done with Core i7 hardware while here are results looking at the equally disturbing performance hits from Core i3 and i5 affected processors too.
Intel's open-source developers working on their security mitigation for the Gen7 graphics hardware have volleyed a new version of the patch series for that mitigation currently causing big hits to Ivybridge / Haswell performance.
Disclosed back in November was the Intel Jump Conditional Code Erratum that necessitated updated CPU microcode to mitigate and with that came with a nearly across the board performance impact. But Intel developers had been working on assembler patches for helping to reduce that performance hit. The GNU Assembler patches were merged back in December while now ahead of LLVM 10.0 that alternative toolchain has an option for helping to recover some of the lost performance.
Intel's open-source graphics driver crew has submitted a final batch of updates to DRM-Next ahead of the Linux 5.6 kernel merge window. The DRM-Next cut-off is this week ahead of the Linux 5.6 window opening up at the start of February.
Coming to Linux last year with the 5.3 kernel was Intel Speed Select Technology support as a Cascade Lake feature for optimizing the per-core performance configurations to favor certain cores at the cost of reducing the performance capacity for other CPU cores. That Intel Speed Select (SST) support for Linux is now being enhanced with core-power controls.
Earlier today we were first to report on an Intel graphics driver patch mitigating a "Gen9" graphics hardware vulnerability. Details on that new security disclosure are coming to light and it turns out older Intel "Gen" graphics are also affected.
On top of the Intel graphics driver patches back from November for denial of service and privilege escalation bugs, the Linux kernel received a new patch today for "CVE-2019-14615" regarding a possible data disclosure with Gen9 graphics hardware.
Sent out shortly before the holidays was an Intel Uncore Frequency driver for Linux servers. That driver has now been revised with the latest fixes and is looking like it could soon land in mainline for helping latency-sensitive servers.
One of the set of patches for Intel's Linux kernel graphics driver that have been floating around for more than one year is about exposing per-client (process) statistics in how each application is making use of the GPU's render/blitter/video hardware and various insightful statistics related to that. The patches aren't queued for mainline yet but at least a new revision of the work was published.
While Intel has offered good Ice Lake support since before the CPUs were shipping (sans taking a bit longer for the Thunderbolt support as a key lone exception, since resolved), a feature that's been publicly known since 2017 is the Fast Short REP MOV behavior and finally with Linux 5.6 that is being made use of for faster memory movements.
Quietly released over the holidays was Intel's quarterly update to the Intel Media Driver that serves as their modern open-source GPU-accelerated video encode/decode solution for Linux systems.
Following AMD's keynote with announcing the Ryzen 4000 series + RX 5600 XT + Threadripper 3990X, Intel now has up their address from the 2020 Consumer Electronics Show (CES). Here are all the highlights from the perspective of a Linux user.
Some Intel server platforms could see better performance with the Linux 5.6 kernel cycle.
Since this summer Intel open-source engineers have been working on adding ACPI Error Disconnect Recover (EDR) support to the Linux kernel and this week marks the eleventh revision to the kernel support for this new ACPI feature.
With Mesa 20.0 expected to ship the "Iris" Gallium3D driver as the default Intel OpenGL Linux driver for Broadwell hardware and newer, I've been ramping up my testing of this open-source driver in recent weeks. For adding to the various generations of CPUs tested, here are some numbers of the latest code when using the UHD Graphics 630 off the high-end Core i9 9900KS processor.
We've been covering Intel's Tiger Lake hardware enablement for Linux since the early patches were posted this summer and that quickly followed with
Intel's open-source graphics driver team responsible for their kernel graphics driver (the i915 Direct Rendering Manager driver) have sent out their first (big) batch of new material to DRM-Next for collection ahead of the Linux 5.6 merge window opening in just over one month's time.
The LLVM Clang compiler continues becoming increasing competitive against the long-standing GNU Compiler Collection (GCC) on Linux x86_64 systems... With tests done on Intel Ice Lake using the Core i7-1065G7, the Clang 9.0 stable performance is delivering over 95% the performance of GCC 9 stable based on over 40 C/C++ benchmarks.
On Friday Intel released SVT-AV1 0.8 with more AVX2/AVX-512 optimizations for this one of the fastest CPU-based AV1 open-source video encoders (and growing decoding support too). Here are some benchmarks of SVT-AV1 0.8 compared to the previous v0.7 release on various Intel and AMD systems.
Disclosed back in November was the Intel Jump Conditional Code (JCC) erratum affecting Skylake and newer CPUs that could lead to "unpredictable behavior" when jump instructions cross cache lines. Intel issued a CPU microcode update to address the problem at a performance cost, but with some compiler toolchain magic, it's possible to mitigate a good portion of that impact.
Intel contributions to Wayland/Weston aren't as frequent as years ago, but they continue volleying interesting work to keep pace with their graphics driver and Direct Rendering Manager subsystem advancements. Their latest work is on adding scaling filter support to libweston in order to supporting filters like nearest-neighbor for yielding less blurry outputs when upscaling.
Intel's Scalable Video Technology SVT-AV1 video encoder/decoder for AV1 content has already been the speediest of the various solutions we have tried, but now a new release is available and it looks to be even faster for CPU-based AV1 video encode/decode.
Back in October 2018 came the initial patches for providing per-process GPU usage reporting to be exposed to user-space for interesting metrics akin to the top command or other system monitoring utilities but for detailed GPU statistics. In October that interesting work finally saw a revision but went dark after that and didn't make it into the recent Linux 5.5 merge window. Now a new spin of that code has been sent out for review.
Well, here is some interesting M+A activity a week ahead of Christmas... Intel just announced they are acquiring AI chipmaker start-up Habana Labs.
In their journey towards the Intel Xe GPUs expected to launch initially next year in the form of Ponte Vecchio, just about one month ago Intel posted patches implementing Shared Virtual Memory support for their Linux graphics driver. Those SVM patches have now been revised for further review in potentially making it for Linux 5.6 should everything look good.
While the Mesa 20.0 cycle is quite young and still over one month to go until the feature freeze for this next quarterly installment of these open-source OpenGL/Vulkan Linux drivers, it's quite exciting already with the changes building up. In particular, on the Intel side they are still positioning for the Intel Gallium3D driver to become the new default on hardware of generations Broadwell and newer. Here is a quick look at how the Intel Gallium3D performance is looking compared to their legacy "i965" classic OpenGL driver that is the current default.
Friday marked not only the release of QEMU 4.2 for Linux virtualization but Intel's open-source crew developing the Cloud-Hypervisor as the Rustlang-based VMM built around Linux's KVM and VirtIO interfaces is out with a big feature release.
With the current Mesa 19.3 there is the Intel Gallium3D driver generally performing much better than their "classic" i965 driver and for Mesa 20.0 it looks to only make more ground as it switches over to this driver by default.
Intel's MKL-DNN Deep Neural Network Library (DNNL) that is open-source and catering to deep learning applications like Tensorflow, PyTorch, DeepLearning4J, and others is nearing its version 2.0 release. With DNNL 2.0 is now support for Data Parallel C++ as Intel's new language as part of their oneAPI initiative.
Intel's OpenSWR software rasterizer for OpenGL as an alternative to the likes of LLVMpipe could soon have OpenGL 4.0 and is a step closer to that code thanks to... Microsoft code.
With Intel Jasper Lake graphics support making it as one of the prominent hardware support additions for Linux 5.5, the user-space OpenGL/Vulkan driver support is now found within Mesa 20.0-devel.
Intel Labs announced "Horse Ridge" as a cryogenic control chip to enable more development and testing around full-stack quantum computing systems.
Back at SC19 Intel released a beta of their oneAPI Base Toolkit for software developers to work on performance-optimized, cross-device software. Complementing that initial software beta is now the oneAPI Level 0 Specification.
Reported on earlier this month is the decision by Linux kernel developers to disable HPET for Intel Coffee Lake systems. The High Precision Event Timer was being disabled since on some Coffee Lake systems at least this timer skews when entering the PC10 power state and that makes the time-stamp counter unstable.
For several release cycles already the Linux kernel has supported Intel's 5-level paging for increasing the virtual and physical address space available to systems while for Linux 5.5 the five-level support is being enabled by default.
Intel's crew maintaining the Scalable Video Technology open-source video encoders on Monday issued a new pre-release of SVT-AV1 in an effort to further speed-up AV1 video encoding on CPUs.
Ending out the week is an exciting development in the Intel open-source graphics driver space... Shared Virtual Memory (SVM) support! This is another step towards their upcoming discrete Xe GPUs and ultimately their exciting oneAPI conquest.
The Khronos Group has officially confirmed Intel's new "Iris" Gallium3D driver as being a conformant OpenGL 4.6 implementation.
Wednesday marked the v1.0.2878 update to Intel's "IGC" Graphics Compiler that is used by their graphics hardware compute stack.
The "Data Streaming Accelerator" (DSA) is a new block on future Intel CPUs that hasn't been talked about much publicly... Until now. Intel's open-source crew has begun detailing DSA for future Intel CPUs that will offer high-performance data movement and transformation operations. The Linux driver enablement has begun.
2934 Intel news articles published on Phoronix.