The Xen Project, hosted at The Linux Foundation, today announced the release of Xen Project Hypervisor 4.8. The latest release focuses on advanced embedded use cases, features to support security-first environments and continued advancement in support of ARM®v8-A based servers. Xen Project technology continues to see growth in these environments due to its flexibility, extensibility and customizability.
As the demand for 64-bit ARMv8-A data centers builds, Xen Project continues to lead by delivering advanced ARM server feature support. Xen Project Hypervisor 4.8 provides initial support for ARM server Live Patching. This allows users to apply security fixes to the Xen Project hypervisor without rebooting, providing five-to-nine reliability for ARM servers. The new feature, available as a preview, also supports the needs of security-first embedded uses cases, such as automotive and avionics.
Over the last year, contributors with strong security and embedded backgrounds have joined the Xen Project. Furthering its stronghold in embedded and security, the project now supports GICv2m (an interrupt controller with MSI capabilities), mmio-sram and IO memory regions with special caching requirements.
“New functionality added to the Xen Project for market segments like automotive, aviation, embedded and security have turned out to be valuable building blocks for traditional server virtualization and hyperscale clouds,” said Lars Kurth, chairperson of the Xen Project. “Some of the innovations contributed by vendors from these segments have helped increase performance, scalability and reduced latency for general workloads, while others led to a more flexible and customizable software architecture that benefit all users of the Xen Project hypervisor and positions us well for future growth across all market segments.”
In the 4.8 release, the general purpose Credit2 scheduler is now supported for production use. Compared to the default Credit scheduler, the Credit2 scheduler is more scalable and is better at supporting latency sensitive workloads such as VDI, video and sound delivery, as well as unikernel applications. Credit2 is still based on a general purpose, weighted fair share, scheduling algorithm unlike some of the more specialized Xen Project schedulers such as RTDS and ARINC653.
Major contributions for this release come from ARM, BitDefender, Bosch, Citrix, Freescale, Intel, Linaro, Oracle, Qualcomm, SUSE, Star Lab, the US National Security Agency, Xilinx, Zentific, and a number of universities and individuals.
The following new features and capabilities are available in Xen Project Hypervisor 4.8:
- Support for Xilinx® Zynq® UltraScale+™ MPSoC: In the embedded space, as multi-chip and multi-OS systems consolidate into virtualized Systems on Chips, Xen Project software’s ability to scale down and provide partitioning with low overhead is key to these environments. Xen Project Hypervisor 4.8 comes with support for the Xilinx Zynq UltraScale+ MPSoC, making it much easier for Xilinx customers to integrate Xen into their solution.
- ARM Architecture Updates: Xen Project 4.8 ARM DomU ACPI support is now able to build ARM64 guests with ACPI support, such as Red Hat Enterprise Linux Server for ARM Development Preview (available via Partner Early Access Program). It can also run unmodified Xen on ARM.
The new release supports alternative runtime patching for ARM64, a powerful technology to dynamically adapt the Xen Project hypervisor code at boot time. This enables the hypervisor to apply workarounds for erratas affecting the processor and to apply optimizations specific to a CPU.
- Intel and x86 Feature Support: The latest version of Xen Project hypervisor adds support of Intel® Advanced Vector Extensions 512 (Intel® AVX-512), which is a natural extension to AVX and AVX2. Intel AVX-512 instructions offer higher performance for the most demanding computational tasks. They represent a significant leap to 512-bit SIMD support. This enables processing of twice the number of data elements that AVX/AVX2 can process with a single instruction and four times that of SSE.
This Xen Project release also comes with PVCLOCK_TSC_STABLE_BIT support, which greatly improves user space performance for time related operations. Another x86 feature is CPUID faulting emulations making it possible to make CPUID fault in HVM userspace program without hardware support.
- PVH v2 update: PVH v2 guest (without PCI passthrough support) ABI is also now stabilized. Guest operating system developers can start porting OSes to this mode, which is simpler and gives them all the goodies that hardware and software provide.
Comments from Xen Project Users and Contributors
“The Xen Project Hypervisor is continuing to grow with new contributors, technologies and use-cases and is increasingly being used in market segments like automotive, mobile and IoT, as well as in its traditional cloud, datacenter and VDI use-cases” said James Bulpin, senior director of technology and chief architect of XenServer, Citrix Systems. “It is exciting to see such a breadth of new development come from a very dedicated and talented group of developers and engineers that make up the Xen Project community.”
“Xen Project hypervisor plays an important role in both the future of embedded systems and in the next generation of cloud computing, especially as these systems increase the use of field programmable gate arrays for acceleration,” said Edgar Iglesias, principal engineer at Xilinx. “Xilinx is committed to creating new programmable technology for next generation systems, and we see Xen Project and its community as being instrumental in this process. We want to congratulate the Xen Project community at large for its work with Xen Project 4.8. Its solid software development practices have delivered yet another great Xen Project hypervisor release.”
“Xen is an extremely important project as part of making the deployment of ARM-based servers a reality,” said Thomas Molgaard, director of product management, Business Segment Group, ARM. “Uninterrupted server availability will be critical for computing in safety-sensitive environments such as connected vehicles. Being able to apply a critical hypervisor patch to fix an issue without affecting operations is a significant step forward.”